refactor: replace hardcoded string with ctxkey constant

2024-06-28 17:48:11 +08:00
127 changed files with 1550 additions and 3916 deletions
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@ -36,12 +36,26 @@ jobs:
      # in the next step as well as the next job.
      - name: Test
        run: go test -cover -coverprofile=coverage.txt ./...
-      - uses: codecov/codecov-action@v4
+
+      - name: Archive code coverage results
+        uses: actions/upload-artifact@v4
        with:
-          token: ${{ secrets.CODECOV_TOKEN }}
+          name: code-coverage
+          path: coverage.txt # Make sure to use the same file name you chose for the "-coverprofile" in the "Test" step
+
+  code_coverage:
+    name: "Code coverage report"
+    if: github.event_name == 'pull_request' # Do not run when workflow is triggered by push to main branch
+    runs-on: ubuntu-latest
+    needs: unit_tests # Depends on the artifact uploaded by the "unit_tests" job
+    steps:
+      - uses: fgrosse/go-coverage-report@v1.0.2 # Consider using a Git revision for maximum security
+        with:
+          coverage-artifact-name: "code-coverage" # can be omitted if you used this default value
+          coverage-file-name: "coverage.txt" # can be omitted if you used this default value

  commit_lint:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
-      - uses: wagoid/commitlint-github-action@v6
+      - uses: wagoid/commitlint-github-action@v6
--- a/.github/workflows/docker-image-amd64-en.yml
+++ b/.github/workflows/docker-image-amd64-en.yml
@ -1,4 +1,4 @@
-name: Publish Docker image (English)
+name: Publish Docker image (amd64, English)

 on:
  push:
@ -34,13 +34,6 @@ jobs:
      - name: Translate
        run: |
          python ./i18n/translate.py --repository_path . --json_file_path ./i18n/en.json
-
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v2
-
      - name: Log in to Docker Hub
        uses: docker/login-action@v2
        with:
@ -58,7 +51,6 @@ jobs:
        uses: docker/build-push-action@v3
        with:
          context: .
-          platforms: linux/amd64,linux/arm64
          push: true
          tags: ${{ steps.meta.outputs.tags }}
          labels: ${{ steps.meta.outputs.labels }}
--- a/.github/workflows/docker-image-amd64.yml
+++ b/.github/workflows/docker-image-amd64.yml
@ -0,0 +1,61 @@
+name: Publish Docker image (amd64)
+
+on:
+  push:
+    tags:
+      - 'v*.*.*'
+  workflow_dispatch:
+    inputs:
+      name:
+        description: 'reason'
+        required: false
+jobs:
+  push_to_registries:
+    name: Push Docker image to multiple registries
+    runs-on: ubuntu-latest
+    permissions:
+      packages: write
+      contents: read
+    steps:
+      - name: Check out the repo
+        uses: actions/checkout@v3
+
+      - name: Check repository URL
+        run: |
+          REPO_URL=$(git config --get remote.origin.url)
+          if [[ $REPO_URL == *"pro" ]]; then
+            exit 1
+          fi        
+
+      - name: Save version info
+        run: |
+          git describe --tags > VERSION 
+
+      - name: Log in to Docker Hub
+        uses: docker/login-action@v2
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Log in to the Container registry
+        uses: docker/login-action@v2
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Extract metadata (tags, labels) for Docker
+        id: meta
+        uses: docker/metadata-action@v4
+        with:
+          images: |
+            justsong/one-api
+            ghcr.io/${{ github.repository }}
+
+      - name: Build and push Docker images
+        uses: docker/build-push-action@v3
+        with:
+          context: .
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
--- a/.github/workflows/docker-image-arm64.yml
+++ b/.github/workflows/docker-image-arm64.yml
@ -1,9 +1,10 @@
-name: Publish Docker image
+name: Publish Docker image (arm64)

 on:
  push:
    tags:
      - 'v*.*.*'
+      - '!*-alpha*'
  workflow_dispatch:
    inputs:
      name:
--- a/8
+++ b/8
@ -1,4 +1,4 @@
-FROM --platform=$BUILDPLATFORM node:16 AS builder
+FROM node:16 as builder

 WORKDIR /web
 COPY ./VERSION .
@ -16,9 +16,7 @@ WORKDIR /web/air
 RUN npm install
 RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat VERSION) npm run build

-FROM golang:alpine AS builder2
-
-RUN apk add --no-cache g++
+FROM golang AS builder2

 ENV GO111MODULE=on \
    CGO_ENABLED=1 \
@ -29,7 +27,7 @@ ADD go.mod go.sum ./
 RUN go mod download
 COPY . .
 COPY --from=builder /web/build ./web/build
-RUN go build -trimpath -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)' -extldflags '-static'" -o one-api
+RUN go build -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)' -extldflags '-static'" -o one-api

 FROM alpine

--- a/README.en.md
+++ b/README.en.md
@ -101,7 +101,7 @@ Nginx reference configuration:
 ```
 server{
   server_name openai.justsong.cn;  # Modify your domain name accordingly
-
+   
   location / {
          client_max_body_size  64m;
          proxy_http_version 1.1;
@ -132,12 +132,12 @@ The initial account username is `root` and password is `123456`.
 1. Download the executable file from [GitHub Releases](https://github.com/songquanpeng/one-api/releases/latest) or compile from source:
   ```shell
   git clone https://github.com/songquanpeng/one-api.git
-
+   
   # Build the frontend
   cd one-api/web/default
   npm install
   npm run build
-
+   
   # Build the backend
   cd ../..
   go mod download
@ -245,41 +245,16 @@ If the channel ID is not provided, load balancing will be used to distribute the
    + Example: `LOG_SQL_DSN=root:123456@tcp(localhost:3306)/oneapi-logs`
 5. `FRONTEND_BASE_URL`: When set, the specified frontend address will be used instead of the backend address.
    + Example: `FRONTEND_BASE_URL=https://openai.justsong.cn`
-6. 'MEMORY_CACHE_ENABLED': Enabling memory caching can cause a certain delay in updating user quotas, with optional values of 'true' and 'false'. If not set, it defaults to 'false'.
-7. `SYNC_FREQUENCY`: When set, the system will periodically sync configurations from the database, with the unit in seconds. If not set, no sync will happen.
+6. `SYNC_FREQUENCY`: When set, the system will periodically sync configurations from the database, with the unit in seconds. If not set, no sync will happen.
    + Example: `SYNC_FREQUENCY=60`
-8. `NODE_TYPE`: When set, specifies the node type. Valid values are `master` and `slave`. If not set, it defaults to `master`.
+7. `NODE_TYPE`: When set, specifies the node type. Valid values are `master` and `slave`. If not set, it defaults to `master`.
    + Example: `NODE_TYPE=slave`
-9. `CHANNEL_UPDATE_FREQUENCY`: When set, it periodically updates the channel balances, with the unit in minutes. If not set, no update will happen.
+8. `CHANNEL_UPDATE_FREQUENCY`: When set, it periodically updates the channel balances, with the unit in minutes. If not set, no update will happen.
    + Example: `CHANNEL_UPDATE_FREQUENCY=1440`
-10. `CHANNEL_TEST_FREQUENCY`: When set, it periodically tests the channels, with the unit in minutes. If not set, no test will happen.
+9. `CHANNEL_TEST_FREQUENCY`: When set, it periodically tests the channels, with the unit in minutes. If not set, no test will happen.
    + Example: `CHANNEL_TEST_FREQUENCY=1440`
-11. `POLLING_INTERVAL`: The time interval (in seconds) between requests when updating channel balances and testing channel availability. Default is no interval.
+10. `POLLING_INTERVAL`: The time interval (in seconds) between requests when updating channel balances and testing channel availability. Default is no interval.
    + Example: `POLLING_INTERVAL=5`
-12. `BATCH_UPDATE_ENABLED`: Enabling batch database update aggregation can cause a certain delay in updating user quotas. The optional values are 'true' and 'false', but if not set, it defaults to 'false'.
-    +Example: ` BATCH_UPDATE_ENABLED=true`
-    +If you encounter an issue with too many database connections, you can try enabling this option.
-13. `BATCH_UPDATE_INTERVAL=5`: The time interval for batch updating aggregates, measured in seconds, defaults to '5'.
-    +Example: ` BATCH_UPDATE_INTERVAL=5`
-14. Request frequency limit:
-    + `GLOBAL_API_RATE_LIMIT`: Global API rate limit (excluding relay requests), the maximum number of requests within three minutes per IP, default to 180.
-    + `GLOBAL_WEL_RATE_LIMIT`: Global web speed limit, the maximum number of requests within three minutes per IP, default to 60.
-15. Encoder cache settings:
-    +`TIKTOKEN_CACHE_DIR`: By default, when the program starts, it will download the encoding of some common word elements online, such as' gpt-3.5 turbo '. In some unstable network environments or offline situations, it may cause startup problems. This directory can be configured to cache data and can be migrated to an offline environment.
-    +`DATA_GYM_CACHE_DIR`: Currently, this configuration has the same function as' TIKTOKEN-CACHE-DIR ', but its priority is not as high as it.
-16. `RELAY_TIMEOUT`: Relay timeout setting, measured in seconds, with no default timeout time set.
-17. `RELAY_PROXY`: After setting up, use this proxy to request APIs.
-18. `USER_CONTENT_REQUEST_TIMEOUT`: The timeout period for users to upload and download content, measured in seconds.
-19. `USER_CONTENT_REQUEST_PROXY`: After setting up, use this agent to request content uploaded by users, such as images.
-20. `SQLITE_BUSY_TIMEOUT`: SQLite lock wait timeout setting, measured in milliseconds, default to '3000'.
-21. `GEMINI_SAFETY_SETTING`: Gemini's security settings are set to 'BLOCK-NONE' by default.
-22. `GEMINI_VERSION`: The Gemini version used by the One API, which defaults to 'v1'.
-23. `THE`: The system's theme setting, default to 'default', specific optional values refer to [here] (./web/README. md).
-24. `ENABLE_METRIC`: Whether to disable channels based on request success rate, default not enabled, optional values are 'true' and 'false'.
-25. `METRIC_QUEUE_SIZE`: Request success rate statistics queue size, default to '10'.
-26. `METRIC_SUCCESS_RATE_THRESHOLD`: Request success rate threshold, default to '0.8'.
-27. `INITIAL_ROOT_TOKEN`: If this value is set, a root user token with the value of the environment variable will be automatically created when the system starts for the first time.
-28. `INITIAL_ROOT_ACCESS_TOKEN`: If this value is set, a system management token will be automatically created for the root user with a value of the environment variable when the system starts for the first time.

 ### Command Line Parameters
 1. `--port <port_number>`: Specifies the port number on which the server listens. Defaults to `3000`.
@ -312,9 +287,7 @@ If the channel ID is not provided, load balancing will be used to distribute the
    + Double-check that your interface address and API Key are correct.

 ## Related Projects
-* [FastGPT](https://github.com/labring/FastGPT): Knowledge question answering system based on the LLM
-* [VChart](https://github.com/VisActor/VChart):  More than just a cross-platform charting library, but also an expressive data storyteller.
-* [VMind](https://github.com/VisActor/VMind):  Not just automatic, but also fantastic. Open-source solution for intelligent visualization.
+[FastGPT](https://github.com/labring/FastGPT): Knowledge question answering system based on the LLM

 ## Note
 This project is an open-source project. Please use it in compliance with OpenAI's [Terms of Use](https://openai.com/policies/terms-of-use) and **applicable laws and regulations**. It must not be used for illegal purposes.
--- a/README.md
+++ b/README.md
@ -53,7 +53,7 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用

 > [!NOTE]
 > 本项目为开源项目，使用者必须在遵循 OpenAI 的[使用条款](https://openai.com/policies/terms-of-use)以及**法律法规**的情况下使用，不得用于非法用途。
->
+> 
 > 根据[《生成式人工智能服务管理暂行办法》](http://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm)的要求，请勿对中国地区公众提供一切未经备案的生成式人工智能服务。

 > [!WARNING]
@ -88,8 +88,6 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
   + [x] [Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai/)
   + [x] [DeepL](https://www.deepl.com/)
   + [x] [together.ai](https://www.together.ai/)
-   + [x] [novita.ai](https://www.novita.ai/)
-   + [x] [硅基流动 SiliconCloud](https://siliconflow.cn/siliconcloud)
 2. 支持配置镜像以及众多[第三方代理服务](https://iamazing.cn/page/openai-api-third-party-services)。
 3. 支持通过**负载均衡**的方式访问多个渠道。
 4. 支持 **stream 模式**，可以通过流式传输实现打字机效果。
@ -146,7 +144,7 @@ Nginx 的参考配置：
 ```
 server{
   server_name openai.justsong.cn;  # 请根据实际情况修改你的域名
-
+   
   location / {
          client_max_body_size  64m;
          proxy_http_version 1.1;
@ -191,12 +189,12 @@ docker-compose ps
 1. 从 [GitHub Releases](https://github.com/songquanpeng/one-api/releases/latest) 下载可执行文件或者从源码编译：
   ```shell
   git clone https://github.com/songquanpeng/one-api.git
-
+   
   # 构建前端
   cd one-api/web/default
   npm install
   npm run build
-
+   
   # 构建后端
   cd ../..
   go mod download
@ -252,9 +250,9 @@ docker run --name chatgpt-web -d -p 3002:3002 -e OPENAI_API_BASE_URL=https://ope
 #### QChatGPT - QQ机器人
 项目主页：https://github.com/RockChinQ/QChatGPT

-根据[文档](https://qchatgpt.rockchin.top)完成部署后，在 `data/provider.json`设置`requester.openai-chat-completions.base-url`为 One API 实例地址，并填写 API Key 到 `keys.openai` 组中，设置 `model` 为要使用的模型名称。
+根据文档完成部署后，在`config.py`设置配置项`openai_config`的`reverse_proxy`为 One API 后端地址，设置`api_key`为 One API 生成的key，并在配置项`completion_api_params`的`model`参数设置为 One API 支持的模型名称。

-运行期间可以通过`!model`命令查看、切换可用模型。
+可安装 [Switcher 插件](https://github.com/RockChinQ/Switcher)在运行时切换所使用的模型。

 ### 部署到第三方平台
 <details>
@ -323,7 +321,7 @@ Render 可以直接部署 docker 镜像，不需要 fork 仓库：https://dashbo
 例如对于 OpenAI 的官方库：
 ```bash
 OPENAI_API_KEY="sk-xxxxxx"
-OPENAI_API_BASE="https://<HOST>:<PORT>/v1"
+OPENAI_API_BASE="https://<HOST>:<PORT>/v1" 
 ```

 ```mermaid
@ -371,34 +369,33 @@ graph LR
   + 例子：`NODE_TYPE=slave`
 9. `CHANNEL_UPDATE_FREQUENCY`：设置之后将定期更新渠道余额，单位为分钟，未设置则不进行更新。
   + 例子：`CHANNEL_UPDATE_FREQUENCY=1440`
-10. `CHANNEL_TEST_FREQUENCY`：设置之后将定期检查渠道，单位为分钟，未设置则不进行检查。 
-   +例子：`CHANNEL_TEST_FREQUENCY=1440`
-11. `POLLING_INTERVAL`：批量更新渠道余额以及测试可用性时的请求间隔，单位为秒，默认无间隔。
+10. `CHANNEL_TEST_FREQUENCY`：设置之后将定期检查渠道，单位为分钟，未设置则不进行检查。
+11. 例子：`CHANNEL_TEST_FREQUENCY=1440`
+12. `POLLING_INTERVAL`：批量更新渠道余额以及测试可用性时的请求间隔，单位为秒，默认无间隔。
    + 例子：`POLLING_INTERVAL=5`
-12. `BATCH_UPDATE_ENABLED`：启用数据库批量更新聚合，会导致用户额度的更新存在一定的延迟可选值为 `true` 和 `false`，未设置则默认为 `false`。
+13. `BATCH_UPDATE_ENABLED`：启用数据库批量更新聚合，会导致用户额度的更新存在一定的延迟可选值为 `true` 和 `false`，未设置则默认为 `false`。
    + 例子：`BATCH_UPDATE_ENABLED=true`
    + 如果你遇到了数据库连接数过多的问题，可以尝试启用该选项。
-13. `BATCH_UPDATE_INTERVAL=5`：批量更新聚合的时间间隔，单位为秒，默认为 `5`。
+14. `BATCH_UPDATE_INTERVAL=5`：批量更新聚合的时间间隔，单位为秒，默认为 `5`。
    + 例子：`BATCH_UPDATE_INTERVAL=5`
-14. 请求频率限制：
+15. 请求频率限制：
    + `GLOBAL_API_RATE_LIMIT`：全局 API 速率限制（除中继请求外），单 ip 三分钟内的最大请求数，默认为 `180`。
    + `GLOBAL_WEB_RATE_LIMIT`：全局 Web 速率限制，单 ip 三分钟内的最大请求数，默认为 `60`。
-15. 编码器缓存设置：
+16. 编码器缓存设置：
    + `TIKTOKEN_CACHE_DIR`：默认程序启动时会联网下载一些通用的词元的编码，如：`gpt-3.5-turbo`，在一些网络环境不稳定，或者离线情况，可能会导致启动有问题，可以配置此目录缓存数据，可迁移到离线环境。
    + `DATA_GYM_CACHE_DIR`：目前该配置作用与 `TIKTOKEN_CACHE_DIR` 一致，但是优先级没有它高。
-16. `RELAY_TIMEOUT`：中继超时设置，单位为秒，默认不设置超时时间。
-17. `RELAY_PROXY`：设置后使用该代理来请求 API。
-18. `USER_CONTENT_REQUEST_TIMEOUT`：用户上传内容下载超时时间，单位为秒。
-19. `USER_CONTENT_REQUEST_PROXY`：设置后使用该代理来请求用户上传的内容，例如图片。
-20. `SQLITE_BUSY_TIMEOUT`：SQLite 锁等待超时设置，单位为毫秒，默认 `3000`。
-21. `GEMINI_SAFETY_SETTING`：Gemini 的安全设置，默认 `BLOCK_NONE`。
-22. `GEMINI_VERSION`：One API 所使用的 Gemini 版本，默认为 `v1`。
-23. `THEME`：系统的主题设置，默认为 `default`，具体可选值参考[此处](./web/README.md)。
-24. `ENABLE_METRIC`：是否根据请求成功率禁用渠道，默认不开启，可选值为 `true` 和 `false`。
-25. `METRIC_QUEUE_SIZE`：请求成功率统计队列大小，默认为 `10`。
-26. `METRIC_SUCCESS_RATE_THRESHOLD`：请求成功率阈值，默认为 `0.8`。
-27. `INITIAL_ROOT_TOKEN`：如果设置了该值，则在系统首次启动时会自动创建一个值为该环境变量值的 root 用户令牌。
-28. `INITIAL_ROOT_ACCESS_TOKEN`：如果设置了该值，则在系统首次启动时会自动创建一个值为该环境变量的 root 用户创建系统管理令牌。
+17. `RELAY_TIMEOUT`：中继超时设置，单位为秒，默认不设置超时时间。
+18. `RELAY_PROXY`：设置后使用该代理来请求 API。
+19. `USER_CONTENT_REQUEST_TIMEOUT`：用户上传内容下载超时时间，单位为秒。
+20. `USER_CONTENT_REQUEST_PROXY`：设置后使用该代理来请求用户上传的内容，例如图片。
+21. `SQLITE_BUSY_TIMEOUT`：SQLite 锁等待超时设置，单位为毫秒，默认 `3000`。
+22. `GEMINI_SAFETY_SETTING`：Gemini 的安全设置，默认 `BLOCK_NONE`。
+23. `GEMINI_VERSION`：One API 所使用的 Gemini 版本，默认为 `v1`。
+24. `THEME`：系统的主题设置，默认为 `default`，具体可选值参考[此处](./web/README.md)。
+25. `ENABLE_METRIC`：是否根据请求成功率禁用渠道，默认不开启，可选值为 `true` 和 `false`。
+26. `METRIC_QUEUE_SIZE`：请求成功率统计队列大小，默认为 `10`。
+27. `METRIC_SUCCESS_RATE_THRESHOLD`：请求成功率阈值，默认为 `0.8`。
+28. `INITIAL_ROOT_TOKEN`：如果设置了该值，则在系统首次启动时会自动创建一个值为该环境变量值的 root 用户令牌。

 ### 命令行参数
 1. `--port <port_number>`: 指定服务器监听的端口号，默认为 `3000`。
@ -451,8 +448,6 @@ https://openai.justsong.cn
 ## 相关项目
 * [FastGPT](https://github.com/labring/FastGPT): 基于 LLM 大语言模型的知识库问答系统
 * [ChatGPT Next Web](https://github.com/Yidadaa/ChatGPT-Next-Web):  一键拥有你自己的跨平台 ChatGPT 应用
-* [VChart](https://github.com/VisActor/VChart):  不只是开箱即用的多端图表库，更是生动灵活的数据故事讲述者。
-* [VMind](https://github.com/VisActor/VMind):  不仅自动，还很智能。开源智能可视化解决方案。

 ## 注意

--- a/common/config/config.go
+++ b/common/config/config.go
@ -35,7 +35,6 @@ var PasswordLoginEnabled = true
 var PasswordRegisterEnabled = true
 var EmailVerificationEnabled = false
 var GitHubOAuthEnabled = false
-var OidcEnabled = false
 var WeChatAuthEnabled = false
 var TurnstileCheckEnabled = false
 var RegisterEnabled = true
@ -71,13 +70,6 @@ var GitHubClientSecret = ""
 var LarkClientId = ""
 var LarkClientSecret = ""

-var OidcClientId = ""
-var OidcClientSecret = ""
-var OidcWellKnown = ""
-var OidcAuthorizationEndpoint = ""
-var OidcTokenEndpoint = ""
-var OidcUserinfoEndpoint = ""
-
 var WeChatServerAddress = ""
 var WeChatServerToken = ""
 var WeChatAccountQRCodeImageURL = ""
@ -151,12 +143,8 @@ var MetricFailChanSize = env.Int("METRIC_FAIL_CHAN_SIZE", 128)

 var InitialRootToken = os.Getenv("INITIAL_ROOT_TOKEN")

-var InitialRootAccessToken = os.Getenv("INITIAL_ROOT_ACCESS_TOKEN")
-
 var GeminiVersion = env.String("GEMINI_VERSION", "v1")

-var OnlyOneLogFile = env.Bool("ONLY_ONE_LOG_FILE", false)
-
 var RelayProxy = env.String("RELAY_PROXY", "")
 var UserContentRequestProxy = env.String("USER_CONTENT_REQUEST_PROXY", "")
 var UserContentRequestTimeout = env.Int("USER_CONTENT_REQUEST_TIMEOUT", 30)
--- a/common/gin.go
+++ b/common/gin.go
@ -31,15 +31,15 @@ func UnmarshalBodyReusable(c *gin.Context, v any) error {
 	contentType := c.Request.Header.Get("Content-Type")
 	if strings.HasPrefix(contentType, "application/json") {
 		err = json.Unmarshal(requestBody, &v)
-		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 	} else {
-		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
-		err = c.ShouldBind(&v)
+		// skip for now
+		// TODO: someday non json request have variant model, we will need to implementation this
 	}
 	if err != nil {
 		return err
 	}
 	// Reset request body
+	c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 	return nil
 }

--- a/common/logger/logger.go
+++ b/common/logger/logger.go
@ -27,12 +27,7 @@ var setupLogOnce sync.Once
 func SetupLogger() {
 	setupLogOnce.Do(func() {
 		if LogDir != "" {
-			var logPath string
-			if config.OnlyOneLogFile {
-				logPath = filepath.Join(LogDir, "oneapi.log")
-			} else {
-				logPath = filepath.Join(LogDir, fmt.Sprintf("oneapi-%s.log", time.Now().Format("20060102")))
-			}
+			logPath := filepath.Join(LogDir, fmt.Sprintf("oneapi-%s.log", time.Now().Format("20060102")))
 			fd, err := os.OpenFile(logPath, os.O_APPEND|os.O_CREATE|os.O_WRONLY, 0644)
 			if err != nil {
 				log.Fatal("failed to open log file")
--- a/common/message/email.go
+++ b/common/message/email.go
@ -6,16 +6,11 @@ import (
 	"encoding/base64"
 	"fmt"
 	"github.com/songquanpeng/one-api/common/config"
-	"net"
 	"net/smtp"
 	"strings"
 	"time"
 )

-func shouldAuth() bool {
-	return config.SMTPAccount != "" || config.SMTPToken != ""
-}
-
 func SendEmail(subject string, receiver string, content string) error {
 	if receiver == "" {
 		return fmt.Errorf("receiver is empty")
@ -46,24 +41,16 @@ func SendEmail(subject string, receiver string, content string) error {
 		"Date: %s\r\n"+
 		"Content-Type: text/html; charset=UTF-8\r\n\r\n%s\r\n",
 		receiver, config.SystemName, config.SMTPFrom, encodedSubject, messageId, time.Now().Format(time.RFC1123Z), content))
-
 	auth := smtp.PlainAuth("", config.SMTPAccount, config.SMTPToken, config.SMTPServer)
 	addr := fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort)
 	to := strings.Split(receiver, ";")

-	if config.SMTPPort == 465 || !shouldAuth() {
-		// need advanced client
-		var conn net.Conn
-		var err error
-		if config.SMTPPort == 465 {
-			tlsConfig := &tls.Config{
-				InsecureSkipVerify: true,
-				ServerName:         config.SMTPServer,
-			}
-			conn, err = tls.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort), tlsConfig)
-		} else {
-			conn, err = net.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort))
+	if config.SMTPPort == 465 {
+		tlsConfig := &tls.Config{
+			InsecureSkipVerify: true,
+			ServerName:         config.SMTPServer,
 		}
+		conn, err := tls.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort), tlsConfig)
 		if err != nil {
 			return err
 		}
@ -72,10 +59,8 @@ func SendEmail(subject string, receiver string, content string) error {
 			return err
 		}
 		defer client.Close()
-		if shouldAuth() {
-			if err = client.Auth(auth); err != nil {
-				return err
-			}
+		if err = client.Auth(auth); err != nil {
+			return err
 		}
 		if err = client.Mail(config.SMTPFrom); err != nil {
 			return err
--- a/common/render/render.go
+++ b/common/render/render.go
@ -1,29 +0,0 @@
-package render
-
-import (
-	"encoding/json"
-	"fmt"
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common"
-	"strings"
-)
-
-func StringData(c *gin.Context, str string) {
-	str = strings.TrimPrefix(str, "data: ")
-	str = strings.TrimSuffix(str, "\r")
-	c.Render(-1, common.CustomEvent{Data: "data: " + str})
-	c.Writer.Flush()
-}
-
-func ObjectData(c *gin.Context, object interface{}) error {
-	jsonData, err := json.Marshal(object)
-	if err != nil {
-		return fmt.Errorf("error marshalling object: %w", err)
-	}
-	StringData(c, string(jsonData))
-	return nil
-}
-
-func Done(c *gin.Context) {
-	StringData(c, "[DONE]")
-}
--- a/controller/auth/oidc.go
+++ b/controller/auth/oidc.go
@ -1,225 +0,0 @@
-package auth
-
-import (
-	"bytes"
-	"encoding/json"
-	"errors"
-	"fmt"
-	"github.com/gin-contrib/sessions"
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/config"
-	"github.com/songquanpeng/one-api/common/logger"
-	"github.com/songquanpeng/one-api/controller"
-	"github.com/songquanpeng/one-api/model"
-	"net/http"
-	"strconv"
-	"time"
-)
-
-type OidcResponse struct {
-	AccessToken  string `json:"access_token"`
-	IDToken      string `json:"id_token"`
-	RefreshToken string `json:"refresh_token"`
-	TokenType    string `json:"token_type"`
-	ExpiresIn    int    `json:"expires_in"`
-	Scope        string `json:"scope"`
-}
-
-type OidcUser struct {
-	OpenID            string `json:"sub"`
-	Email             string `json:"email"`
-	Name              string `json:"name"`
-	PreferredUsername string `json:"preferred_username"`
-	Picture           string `json:"picture"`
-}
-
-func getOidcUserInfoByCode(code string) (*OidcUser, error) {
-	if code == "" {
-		return nil, errors.New("无效的参数")
-	}
-	values := map[string]string{
-		"client_id":     config.OidcClientId,
-		"client_secret": config.OidcClientSecret,
-		"code":          code,
-		"grant_type":    "authorization_code",
-		"redirect_uri":  fmt.Sprintf("%s/oauth/oidc", config.ServerAddress),
-	}
-	jsonData, err := json.Marshal(values)
-	if err != nil {
-		return nil, err
-	}
-	req, err := http.NewRequest("POST", config.OidcTokenEndpoint, bytes.NewBuffer(jsonData))
-	if err != nil {
-		return nil, err
-	}
-	req.Header.Set("Content-Type", "application/json")
-	req.Header.Set("Accept", "application/json")
-	client := http.Client{
-		Timeout: 5 * time.Second,
-	}
-	res, err := client.Do(req)
-	if err != nil {
-		logger.SysLog(err.Error())
-		return nil, errors.New("无法连接至 OIDC 服务器，请稍后重试！")
-	}
-	defer res.Body.Close()
-	var oidcResponse OidcResponse
-	err = json.NewDecoder(res.Body).Decode(&oidcResponse)
-	if err != nil {
-		return nil, err
-	}
-	req, err = http.NewRequest("GET", config.OidcUserinfoEndpoint, nil)
-	if err != nil {
-		return nil, err
-	}
-	req.Header.Set("Authorization", "Bearer "+oidcResponse.AccessToken)
-	res2, err := client.Do(req)
-	if err != nil {
-		logger.SysLog(err.Error())
-		return nil, errors.New("无法连接至 OIDC 服务器，请稍后重试！")
-	}
-	var oidcUser OidcUser
-	err = json.NewDecoder(res2.Body).Decode(&oidcUser)
-	if err != nil {
-		return nil, err
-	}
-	return &oidcUser, nil
-}
-
-func OidcAuth(c *gin.Context) {
-	session := sessions.Default(c)
-	state := c.Query("state")
-	if state == "" || session.Get("oauth_state") == nil || state != session.Get("oauth_state").(string) {
-		c.JSON(http.StatusForbidden, gin.H{
-			"success": false,
-			"message": "state is empty or not same",
-		})
-		return
-	}
-	username := session.Get("username")
-	if username != nil {
-		OidcBind(c)
-		return
-	}
-	if !config.OidcEnabled {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": "管理员未开启通过 OIDC 登录以及注册",
-		})
-		return
-	}
-	code := c.Query("code")
-	oidcUser, err := getOidcUserInfoByCode(code)
-	if err != nil {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": err.Error(),
-		})
-		return
-	}
-	user := model.User{
-		OidcId: oidcUser.OpenID,
-	}
-	if model.IsOidcIdAlreadyTaken(user.OidcId) {
-		err := user.FillUserByOidcId()
-		if err != nil {
-			c.JSON(http.StatusOK, gin.H{
-				"success": false,
-				"message": err.Error(),
-			})
-			return
-		}
-	} else {
-		if config.RegisterEnabled {
-			user.Email = oidcUser.Email
-			if oidcUser.PreferredUsername != "" {
-				user.Username = oidcUser.PreferredUsername
-			} else {
-				user.Username = "oidc_" + strconv.Itoa(model.GetMaxUserId()+1)
-			}
-			if oidcUser.Name != "" {
-				user.DisplayName = oidcUser.Name
-			} else {
-				user.DisplayName = "OIDC User"
-			}
-			err := user.Insert(0)
-			if err != nil {
-				c.JSON(http.StatusOK, gin.H{
-					"success": false,
-					"message": err.Error(),
-				})
-				return
-			}
-		} else {
-			c.JSON(http.StatusOK, gin.H{
-				"success": false,
-				"message": "管理员关闭了新用户注册",
-			})
-			return
-		}
-	}
-
-	if user.Status != model.UserStatusEnabled {
-		c.JSON(http.StatusOK, gin.H{
-			"message": "用户已被封禁",
-			"success": false,
-		})
-		return
-	}
-	controller.SetupLogin(&user, c)
-}
-
-func OidcBind(c *gin.Context) {
-	if !config.OidcEnabled {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": "管理员未开启通过 OIDC 登录以及注册",
-		})
-		return
-	}
-	code := c.Query("code")
-	oidcUser, err := getOidcUserInfoByCode(code)
-	if err != nil {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": err.Error(),
-		})
-		return
-	}
-	user := model.User{
-		OidcId: oidcUser.OpenID,
-	}
-	if model.IsOidcIdAlreadyTaken(user.OidcId) {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": "该 OIDC 账户已被绑定",
-		})
-		return
-	}
-	session := sessions.Default(c)
-	id := session.Get("id")
-	// id := c.GetInt("id")  // critical bug!
-	user.Id = id.(int)
-	err = user.FillUserById()
-	if err != nil {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": err.Error(),
-		})
-		return
-	}
-	user.OidcId = oidcUser.OpenID
-	err = user.Update(false)
-	if err != nil {
-		c.JSON(http.StatusOK, gin.H{
-			"success": false,
-			"message": err.Error(),
-		})
-		return
-	}
-	c.JSON(http.StatusOK, gin.H{
-		"success": true,
-		"message": "bind",
-	})
-	return
-}
--- a/controller/billing.go
+++ b/controller/billing.go
@ -17,11 +17,9 @@ func GetSubscription(c *gin.Context) {
 	if config.DisplayTokenStatEnabled {
 		tokenId := c.GetInt(ctxkey.TokenId)
 		token, err = model.GetTokenById(tokenId)
-		if err == nil {
-			expiredTime = token.ExpiredTime
-			remainQuota = token.RemainQuota
-			usedQuota = token.UsedQuota
-		}
+		expiredTime = token.ExpiredTime
+		remainQuota = token.RemainQuota
+		usedQuota = token.UsedQuota
 	} else {
 		userId := c.GetInt(ctxkey.Id)
 		remainQuota, err = model.GetUserQuota(userId)
--- a/controller/channel-billing.go
+++ b/controller/channel-billing.go
@ -81,26 +81,6 @@ type APGC2DGPTUsageResponse struct {
 	TotalUsed      float64 `json:"total_used"`
 }

-type SiliconFlowUsageResponse struct {
-	Code    int    `json:"code"`
-	Message string `json:"message"`
-	Status  bool   `json:"status"`
-	Data    struct {
-		ID            string `json:"id"`
-		Name          string `json:"name"`
-		Image         string `json:"image"`
-		Email         string `json:"email"`
-		IsAdmin       bool   `json:"isAdmin"`
-		Balance       string `json:"balance"`
-		Status        string `json:"status"`
-		Introduction  string `json:"introduction"`
-		Role          string `json:"role"`
-		ChargeBalance string `json:"chargeBalance"`
-		TotalBalance  string `json:"totalBalance"`
-		Category      string `json:"category"`
-	} `json:"data"`
-}
-
 // GetAuthHeader get auth header
 func GetAuthHeader(token string) http.Header {
 	h := http.Header{}
@ -223,28 +203,6 @@ func updateChannelAIGC2DBalance(channel *model.Channel) (float64, error) {
 	return response.TotalAvailable, nil
 }

-func updateChannelSiliconFlowBalance(channel *model.Channel) (float64, error) {
-	url := "https://api.siliconflow.cn/v1/user/info"
-	body, err := GetResponseBody("GET", url, channel, GetAuthHeader(channel.Key))
-	if err != nil {
-		return 0, err
-	}
-	response := SiliconFlowUsageResponse{}
-	err = json.Unmarshal(body, &response)
-	if err != nil {
-		return 0, err
-	}
-	if response.Code != 20000 {
-		return 0, fmt.Errorf("code: %d, message: %s", response.Code, response.Message)
-	}
-	balance, err := strconv.ParseFloat(response.Data.Balance, 64)
-	if err != nil {
-		return 0, err
-	}
-	channel.UpdateBalance(balance)
-	return balance, nil
-}
-
 func updateChannelBalance(channel *model.Channel) (float64, error) {
 	baseURL := channeltype.ChannelBaseURLs[channel.Type]
 	if channel.GetBaseURL() == "" {
@ -269,8 +227,6 @@ func updateChannelBalance(channel *model.Channel) (float64, error) {
 		return updateChannelAPI2GPTBalance(channel)
 	case channeltype.AIGC2D:
 		return updateChannelAIGC2DBalance(channel)
-	case channeltype.SiliconFlow:
-		return updateChannelSiliconFlowBalance(channel)
 	default:
 		return 0, errors.New("尚未实现")
 	}
--- a/controller/channel-test.go
+++ b/controller/channel-test.go
@ -14,7 +14,6 @@ import (
 	"sync"
 	"time"

-	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/common/logger"
@ -28,15 +27,15 @@ import (
 	"github.com/songquanpeng/one-api/relay/meta"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
+
+	"github.com/gin-gonic/gin"
 )

-func buildTestRequest(model string) *relaymodel.GeneralOpenAIRequest {
-	if model == "" {
-		model = "gpt-3.5-turbo"
-	}
+func buildTestRequest() *relaymodel.GeneralOpenAIRequest {
 	testRequest := &relaymodel.GeneralOpenAIRequest{
 		MaxTokens: 2,
-		Model:     model,
+		Stream:    false,
+		Model:     "gpt-3.5-turbo",
 	}
 	testMessage := relaymodel.Message{
 		Role:    "user",
@ -46,7 +45,7 @@ func buildTestRequest(model string) *relaymodel.GeneralOpenAIRequest {
 	return testRequest
 }

-func testChannel(channel *model.Channel, request *relaymodel.GeneralOpenAIRequest) (err error, openaiErr *relaymodel.Error) {
+func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error) {
 	w := httptest.NewRecorder()
 	c, _ := gin.CreateTestContext(w)
 	c.Request = &http.Request{
@ -69,8 +68,12 @@ func testChannel(channel *model.Channel, request *relaymodel.GeneralOpenAIReques
 		return fmt.Errorf("invalid api type: %d, adaptor is nil", apiType), nil
 	}
 	adaptor.Init(meta)
-	modelName := request.Model
+	var modelName string
+	modelList := adaptor.GetModelList()
 	modelMap := channel.GetModelMapping()
+	if len(modelList) != 0 {
+		modelName = modelList[0]
+	}
 	if modelName == "" || !strings.Contains(channel.Models, modelName) {
 		modelNames := strings.Split(channel.Models, ",")
 		if len(modelNames) > 0 {
@ -80,8 +83,9 @@ func testChannel(channel *model.Channel, request *relaymodel.GeneralOpenAIReques
 			modelName = modelMap[modelName]
 		}
 	}
-	meta.OriginModelName, meta.ActualModelName = request.Model, modelName
+	request := buildTestRequest()
 	request.Model = modelName
+	meta.OriginModelName, meta.ActualModelName = modelName, modelName
 	convertedRequest, err := adaptor.ConvertRequest(c, relaymode.ChatCompletions, request)
 	if err != nil {
 		return err, nil
@ -135,15 +139,10 @@ func TestChannel(c *gin.Context) {
 		})
 		return
 	}
-	model := c.Query("model")
-	testRequest := buildTestRequest(model)
 	tik := time.Now()
-	err, _ = testChannel(channel, testRequest)
+	err, _ = testChannel(channel)
 	tok := time.Now()
 	milliseconds := tok.Sub(tik).Milliseconds()
-	if err != nil {
-		milliseconds = 0
-	}
 	go channel.UpdateResponseTime(milliseconds)
 	consumedTime := float64(milliseconds) / 1000.0
 	if err != nil {
@ -151,7 +150,6 @@ func TestChannel(c *gin.Context) {
 			"success": false,
 			"message": err.Error(),
 			"time":    consumedTime,
-			"model":   model,
 		})
 		return
 	}
@ -159,7 +157,6 @@ func TestChannel(c *gin.Context) {
 		"success": true,
 		"message": "",
 		"time":    consumedTime,
-		"model":   model,
 	})
 	return
 }
@ -190,12 +187,11 @@ func testChannels(notify bool, scope string) error {
 		for _, channel := range channels {
 			isChannelEnabled := channel.Status == model.ChannelStatusEnabled
 			tik := time.Now()
-			testRequest := buildTestRequest("")
-			err, openaiErr := testChannel(channel, testRequest)
+			err, openaiErr := testChannel(channel)
 			tok := time.Now()
 			milliseconds := tok.Sub(tik).Milliseconds()
 			if isChannelEnabled && milliseconds > disableThreshold {
-				err = fmt.Errorf("响应时间 %.2fs 超过阈值 %.2fs", float64(milliseconds)/1000.0, float64(disableThreshold)/1000.0)
+				err = errors.New(fmt.Sprintf("响应时间 %.2fs 超过阈值 %.2fs", float64(milliseconds)/1000.0, float64(disableThreshold)/1000.0))
 				if config.AutomaticDisableChannelEnabled {
 					monitor.DisableChannel(channel.Id, channel.Name, err.Error())
 				} else {
--- a/controller/misc.go
+++ b/controller/misc.go
@ -18,30 +18,24 @@ func GetStatus(c *gin.Context) {
 		"success": true,
 		"message": "",
 		"data": gin.H{
-			"version":                     common.Version,
-			"start_time":                  common.StartTime,
-			"email_verification":          config.EmailVerificationEnabled,
-			"github_oauth":                config.GitHubOAuthEnabled,
-			"github_client_id":            config.GitHubClientId,
-			"lark_client_id":              config.LarkClientId,
-			"system_name":                 config.SystemName,
-			"logo":                        config.Logo,
-			"footer_html":                 config.Footer,
-			"wechat_qrcode":               config.WeChatAccountQRCodeImageURL,
-			"wechat_login":                config.WeChatAuthEnabled,
-			"server_address":              config.ServerAddress,
-			"turnstile_check":             config.TurnstileCheckEnabled,
-			"turnstile_site_key":          config.TurnstileSiteKey,
-			"top_up_link":                 config.TopUpLink,
-			"chat_link":                   config.ChatLink,
-			"quota_per_unit":              config.QuotaPerUnit,
-			"display_in_currency":         config.DisplayInCurrencyEnabled,
-			"oidc":                        config.OidcEnabled,
-			"oidc_client_id":              config.OidcClientId,
-			"oidc_well_known":             config.OidcWellKnown,
-			"oidc_authorization_endpoint": config.OidcAuthorizationEndpoint,
-			"oidc_token_endpoint":         config.OidcTokenEndpoint,
-			"oidc_userinfo_endpoint":      config.OidcUserinfoEndpoint,
+			"version":             common.Version,
+			"start_time":          common.StartTime,
+			"email_verification":  config.EmailVerificationEnabled,
+			"github_oauth":        config.GitHubOAuthEnabled,
+			"github_client_id":    config.GitHubClientId,
+			"lark_client_id":      config.LarkClientId,
+			"system_name":         config.SystemName,
+			"logo":                config.Logo,
+			"footer_html":         config.Footer,
+			"wechat_qrcode":       config.WeChatAccountQRCodeImageURL,
+			"wechat_login":        config.WeChatAuthEnabled,
+			"server_address":      config.ServerAddress,
+			"turnstile_check":     config.TurnstileCheckEnabled,
+			"turnstile_site_key":  config.TurnstileSiteKey,
+			"top_up_link":         config.TopUpLink,
+			"chat_link":           config.ChatLink,
+			"quota_per_unit":      config.QuotaPerUnit,
+			"display_in_currency": config.DisplayInCurrencyEnabled,
 		},
 	})
 	return
--- a/controller/relay.go
+++ b/controller/relay.go
@ -34,8 +34,6 @@ func relayHelper(c *gin.Context, relayMode int) *model.ErrorWithStatusCode {
 		fallthrough
 	case relaymode.AudioTranscription:
 		err = controller.RelayAudioHelper(c, relayMode)
-	case relaymode.Proxy:
-		err = controller.RelayProxyHelper(c, relayMode)
 	default:
 		err = controller.RelayTextHelper(c)
 	}
@ -87,15 +85,12 @@ func Relay(c *gin.Context) {
 		channelId := c.GetInt(ctxkey.ChannelId)
 		lastFailedChannelId = channelId
 		channelName := c.GetString(ctxkey.ChannelName)
-		// BUG: bizErr is in race condition
 		go processChannelRelayError(ctx, userId, channelId, channelName, bizErr)
 	}
 	if bizErr != nil {
 		if bizErr.StatusCode == http.StatusTooManyRequests {
 			bizErr.Error.Message = "当前分组上游负载已饱和，请稍后再试"
 		}
-
-		// BUG: bizErr is in race condition
 		bizErr.Error.Message = helper.MessageWithRequestId(bizErr.Error.Message, requestId)
 		c.JSON(bizErr.StatusCode, gin.H{
 			"error": bizErr.Error,
--- a/go.mod
+++ b/go.mod
@ -4,7 +4,6 @@ module github.com/songquanpeng/one-api
 go 1.20

 require (
-	cloud.google.com/go/iam v1.1.10
 	github.com/aws/aws-sdk-go-v2 v1.27.0
 	github.com/aws/aws-sdk-go-v2/credentials v1.17.15
 	github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.8.3
@ -20,14 +19,12 @@ require (
 	github.com/gorilla/websocket v1.5.1
 	github.com/jinzhu/copier v0.4.0
 	github.com/joho/godotenv v1.5.1
-	github.com/patrickmn/go-cache v2.1.0+incompatible
 	github.com/pkg/errors v0.9.1
 	github.com/pkoukk/tiktoken-go v0.1.7
 	github.com/smartystreets/goconvey v1.8.1
 	github.com/stretchr/testify v1.9.0
-	golang.org/x/crypto v0.24.0
-	golang.org/x/image v0.18.0
-	google.golang.org/api v0.187.0
+	golang.org/x/crypto v0.23.0
+	golang.org/x/image v0.16.0
 	gorm.io/driver/mysql v1.5.6
 	gorm.io/driver/postgres v1.5.7
 	gorm.io/driver/sqlite v1.5.5
@ -35,9 +32,6 @@ require (
 )

 require (
-	cloud.google.com/go/auth v0.6.1 // indirect
-	cloud.google.com/go/auth/oauth2adapt v0.2.2 // indirect
-	cloud.google.com/go/compute/metadata v0.3.0 // indirect
 	filippo.io/edwards25519 v1.1.0 // indirect
 	github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2 // indirect
 	github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7 // indirect
@ -51,21 +45,13 @@ require (
 	github.com/davecgh/go-spew v1.1.1 // indirect
 	github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f // indirect
 	github.com/dlclark/regexp2 v1.11.0 // indirect
-	github.com/felixge/httpsnoop v1.0.4 // indirect
 	github.com/fsnotify/fsnotify v1.7.0 // indirect
 	github.com/gabriel-vasile/mimetype v1.4.3 // indirect
 	github.com/gin-contrib/sse v0.1.0 // indirect
-	github.com/go-logr/logr v1.4.1 // indirect
-	github.com/go-logr/stdr v1.2.2 // indirect
 	github.com/go-playground/locales v0.14.1 // indirect
 	github.com/go-playground/universal-translator v0.18.1 // indirect
 	github.com/go-sql-driver/mysql v1.8.1 // indirect
 	github.com/goccy/go-json v0.10.3 // indirect
-	github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da // indirect
-	github.com/golang/protobuf v1.5.4 // indirect
-	github.com/google/s2a-go v0.1.7 // indirect
-	github.com/googleapis/enterprise-certificate-proxy v0.3.2 // indirect
-	github.com/googleapis/gax-go/v2 v2.12.5 // indirect
 	github.com/gopherjs/gopherjs v1.17.2 // indirect
 	github.com/gorilla/context v1.1.2 // indirect
 	github.com/gorilla/securecookie v1.1.2 // indirect
@ -82,7 +68,7 @@ require (
 	github.com/kr/text v0.2.0 // indirect
 	github.com/leodido/go-urn v1.4.0 // indirect
 	github.com/mattn/go-isatty v0.0.20 // indirect
-	github.com/mattn/go-sqlite3 v1.14.22 // indirect
+	github.com/mattn/go-sqlite3 v2.0.3+incompatible // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/pelletier/go-toml/v2 v2.2.2 // indirect
@ -90,22 +76,11 @@ require (
 	github.com/smarty/assertions v1.15.0 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	github.com/ugorji/go/codec v1.2.12 // indirect
-	go.opencensus.io v0.24.0 // indirect
-	go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.49.0 // indirect
-	go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.49.0 // indirect
-	go.opentelemetry.io/otel v1.24.0 // indirect
-	go.opentelemetry.io/otel/metric v1.24.0 // indirect
-	go.opentelemetry.io/otel/trace v1.24.0 // indirect
 	golang.org/x/arch v0.8.0 // indirect
-	golang.org/x/net v0.26.0 // indirect
-	golang.org/x/oauth2 v0.21.0 // indirect
+	golang.org/x/net v0.25.0 // indirect
 	golang.org/x/sync v0.7.0 // indirect
-	golang.org/x/sys v0.21.0 // indirect
-	golang.org/x/text v0.16.0 // indirect
-	golang.org/x/time v0.5.0 // indirect
-	google.golang.org/genproto/googleapis/api v0.0.0-20240617180043-68d350f18fd4 // indirect
-	google.golang.org/genproto/googleapis/rpc v0.0.0-20240624140628-dc46fd24d27d // indirect
-	google.golang.org/grpc v1.64.1 // indirect
-	google.golang.org/protobuf v1.34.2 // indirect
+	golang.org/x/sys v0.20.0 // indirect
+	golang.org/x/text v0.15.0 // indirect
+	google.golang.org/protobuf v1.34.1 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 )
--- a/go.sum
+++ b/go.sum
@ -1,15 +1,5 @@
-cloud.google.com/go v0.26.0/go.mod h1:aQUYkXzVsufM+DwF1aE+0xfcU+56JwCaLick0ClmMTw=
-cloud.google.com/go/auth v0.6.1 h1:T0Zw1XM5c1GlpN2HYr2s+m3vr1p2wy+8VN+Z1FKxW38=
-cloud.google.com/go/auth v0.6.1/go.mod h1:eFHG7zDzbXHKmjJddFG/rBlcGp6t25SwRUiEQSlO4x4=
-cloud.google.com/go/auth/oauth2adapt v0.2.2 h1:+TTV8aXpjeChS9M+aTtN/TjdQnzJvmzKFt//oWu7HX4=
-cloud.google.com/go/auth/oauth2adapt v0.2.2/go.mod h1:wcYjgpZI9+Yu7LyYBg4pqSiaRkfEK3GQcpb7C/uyF1Q=
-cloud.google.com/go/compute/metadata v0.3.0 h1:Tz+eQXMEqDIKRsmY3cHTL6FVaynIjX2QxYC4trgAKZc=
-cloud.google.com/go/compute/metadata v0.3.0/go.mod h1:zFmK7XCadkQkj6TtorcaGlCW1hT1fIilQDwofLpJ20k=
-cloud.google.com/go/iam v1.1.10 h1:ZSAr64oEhQSClwBL670MsJAW5/RLiC6kfw3Bqmd5ZDI=
-cloud.google.com/go/iam v1.1.10/go.mod h1:iEgMq62sg8zx446GCaijmA2Miwg5o3UbO+nI47WHJps=
 filippo.io/edwards25519 v1.1.0 h1:FNf4tywRC1HmFuKW5xopWpigGjJKiJSV0Cqo0cJWDaA=
 filippo.io/edwards25519 v1.1.0/go.mod h1:BxyFTGdWcka3PhytdK4V28tE5sGfRvvvRV7EaN4VDT4=
-github.com/BurntSushi/toml v0.3.1/go.mod h1:xHWCNGjB5oqiDr8zfno3MHue2Ht5sIBksp03qcyfWMU=
 github.com/aws/aws-sdk-go-v2 v1.27.0 h1:7bZWKoXhzI+mMR/HjdMx8ZCC5+6fY0lS5tr0bbgiLlo=
 github.com/aws/aws-sdk-go-v2 v1.27.0/go.mod h1:ffIFB97e2yNsv4aTSGkqtHnppsIJzw7G7BReUZ3jCXM=
 github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2 h1:x6xsQXGSmW6frevwDA+vi/wqhp1ct18mVXYN08/93to=
@ -28,15 +18,12 @@ github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc
 github.com/bytedance/sonic v1.11.6/go.mod h1:LysEHSvpvDySVdC2f87zGWf6CIKJcAvqab1ZaiQtds4=
 github.com/bytedance/sonic/loader v0.1.1 h1:c+e5Pt1k/cy5wMveRDyk2X4B9hF4g7an8N3zCYjJFNM=
 github.com/bytedance/sonic/loader v0.1.1/go.mod h1:ncP89zfokxS5LZrJxl5z0UJcsk4M4yY2JpfqGeCtNLU=
-github.com/census-instrumentation/opencensus-proto v0.2.1/go.mod h1:f6KPmirojxKA12rnyqOA5BBL4O983OfeGPqjHWSTneU=
 github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs=
 github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
-github.com/client9/misspell v0.3.4/go.mod h1:qj6jICC3Q7zFZvVWo7KLAzC3yx5G7kyvSDkc90ppPyw=
 github.com/cloudwego/base64x v0.1.4 h1:jwCgWpFanWmN8xoIUHa2rtzmkd5J2plF/dnLS6Xd/0Y=
 github.com/cloudwego/base64x v0.1.4/go.mod h1:0zlkT4Wn5C6NdauXdJRhSKRlJvmclQ1hhJgA0rcu/8w=
 github.com/cloudwego/iasm v0.2.0 h1:1KNIy1I1H9hNNFEEH3DVnI4UujN+1zjpuk6gwHLTssg=
 github.com/cloudwego/iasm v0.2.0/go.mod h1:8rXZaNYT2n95jn+zTI1sDr+IgcD2GVs0nlbbQPiEFhY=
-github.com/cncf/udpa/go v0.0.0-20191209042840-269d4d468f6f/go.mod h1:M8M6+tZqaGXZJjfX53e64911xZQV5JYwmTeXPW+k8Sc=
 github.com/creack/pty v1.1.9/go.mod h1:oKZEueFk5CKHvIhNR5MUki03XCEU+Q6VDXinZuGJ33E=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
@ -45,12 +32,6 @@ github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f h1:lO4WD4F/r
 github.com/dgryski/go-rendezvous v0.0.0-20200823014737-9f7001d12a5f/go.mod h1:cuUVRXasLTGF7a8hSLbxyZXjz+1KgoB3wDUb6vlszIc=
 github.com/dlclark/regexp2 v1.11.0 h1:G/nrcoOa7ZXlpoa/91N3X7mM3r8eIlMBBJZvsz/mxKI=
 github.com/dlclark/regexp2 v1.11.0/go.mod h1:DHkYz0B9wPfa6wondMfaivmHpzrQ3v9q8cnmRbL6yW8=
-github.com/envoyproxy/go-control-plane v0.9.0/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
-github.com/envoyproxy/go-control-plane v0.9.1-0.20191026205805-5f8ba28d4473/go.mod h1:YTl/9mNaCwkRvm6d1a2C3ymFceY/DCBVvsKhRF0iEA4=
-github.com/envoyproxy/go-control-plane v0.9.4/go.mod h1:6rpuAdCZL397s3pYoYcLgu1mIlRU8Am5FuJP05cCM98=
-github.com/envoyproxy/protoc-gen-validate v0.1.0/go.mod h1:iSmxcyjqTsJpI2R4NaDN7+kN2VEUnK/pcBlmesArF7c=
-github.com/felixge/httpsnoop v1.0.4 h1:NFTV2Zj1bL4mc9sqWACXbQFVBBg2W3GPvqp8/ESS2Wg=
-github.com/felixge/httpsnoop v1.0.4/go.mod h1:m8KPJKqk1gH5J9DgRY2ASl2lWCfGKXixSwevea8zH2U=
 github.com/fsnotify/fsnotify v1.7.0 h1:8JEhPFa5W2WU7YfeZzPNqzMP6Lwt7L2715Ggo0nosvA=
 github.com/fsnotify/fsnotify v1.7.0/go.mod h1:40Bi/Hjc2AVfZrqy+aj+yEI+/bRxZnMJyTJwOpGvigM=
 github.com/gabriel-vasile/mimetype v1.4.3 h1:in2uUcidCuFcDKtdcBxlR0rJ1+fsokWf+uqxgUFjbI0=
@ -67,11 +48,6 @@ github.com/gin-contrib/static v1.1.2 h1:c3kT4bFkUJn2aoRU3s6XnMjJT8J6nNWJkR0Nglqm
 github.com/gin-contrib/static v1.1.2/go.mod h1:Fw90ozjHCmZBWbgrsqrDvO28YbhKEKzKp8GixhR4yLw=
 github.com/gin-gonic/gin v1.10.0 h1:nTuyha1TYqgedzytsKYqna+DfLos46nTv2ygFy86HFU=
 github.com/gin-gonic/gin v1.10.0/go.mod h1:4PMNQiOhvDRa013RKVbsiNwoyezlm2rm0uX/T7kzp5Y=
-github.com/go-logr/logr v1.2.2/go.mod h1:jdQByPbusPIv2/zmleS9BjJVeZ6kBagPoEUsqbVz/1A=
-github.com/go-logr/logr v1.4.1 h1:pKouT5E8xu9zeFC39JXRDukb6JFQPXM5p5I91188VAQ=
-github.com/go-logr/logr v1.4.1/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
-github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
-github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
 github.com/go-playground/assert/v2 v2.2.0 h1:JvknZsQTYeFEAhQwI4qEt9cyV5ONwRHC+lYKSsYSR8s=
 github.com/go-playground/locales v0.14.1 h1:EWaQ/wswjilfKLTECiXz7Rh+3BjFhfDFKv/oXslEjJA=
 github.com/go-playground/locales v0.14.1/go.mod h1:hxrqLVvrK65+Rwrd5Fc6F2O76J/NuW9t0sjnWqG1slY=
@ -88,40 +64,11 @@ github.com/goccy/go-json v0.10.3 h1:KZ5WoDbxAIgm2HNbYckL0se1fHD6rz5j4ywS6ebzDqA=
 github.com/goccy/go-json v0.10.3/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M=
 github.com/golang-jwt/jwt v3.2.2+incompatible h1:IfV12K8xAKAnZqdXVzCZ+TOjboZ2keLg81eXfW3O+oY=
 github.com/golang-jwt/jwt v3.2.2+incompatible/go.mod h1:8pz2t5EyA70fFQQSrl6XZXzqecmYZeUEB8OUGHkxJ+I=
-github.com/golang/glog v0.0.0-20160126235308-23def4e6c14b/go.mod h1:SBH7ygxi8pfUlaOkMMuAQtPIUF8ecWP5IEl/CR7VP2Q=
-github.com/golang/groupcache v0.0.0-20200121045136-8c9f03a8e57e/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
-github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da h1:oI5xCqsCo564l8iNU+DwB5epxmsaqB+rhGL0m5jtYqE=
-github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
-github.com/golang/mock v1.1.1/go.mod h1:oTYuIxOrZwtPieC+H1uAHpcLFnEyAGVDL/k47Jfbm0A=
-github.com/golang/protobuf v1.2.0/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
-github.com/golang/protobuf v1.3.2/go.mod h1:6lQm79b+lXiMfvg/cZm0SGofjICqVBUtrP5yJMmIC1U=
-github.com/golang/protobuf v1.4.0-rc.1/go.mod h1:ceaxUfeHdC40wWswd/P6IGgMaK3YpKi5j83Wpe3EHw8=
-github.com/golang/protobuf v1.4.0-rc.1.0.20200221234624-67d41d38c208/go.mod h1:xKAWHe0F5eneWXFV3EuXVDTCmh+JuBKY0li0aMyXATA=
-github.com/golang/protobuf v1.4.0-rc.2/go.mod h1:LlEzMj4AhA7rCAGe4KMBDvJI+AwstrUpVNzEA03Pprs=
-github.com/golang/protobuf v1.4.0-rc.4.0.20200313231945-b860323f09d0/go.mod h1:WU3c8KckQ9AFe+yFwt9sWVRKCVIyN9cPHBJSNnbL67w=
-github.com/golang/protobuf v1.4.0/go.mod h1:jodUvKwWbYaEsadDk5Fwe5c77LiNKVO9IDvqG2KuDX0=
-github.com/golang/protobuf v1.4.1/go.mod h1:U8fpvMrcmy5pZrNK1lt4xCsGvpyWQ/VVv6QDs8UjoX8=
-github.com/golang/protobuf v1.4.3/go.mod h1:oDoupMAO8OvCJWAcko0GGGIgR6R6ocIYbsSw735rRwI=
-github.com/golang/protobuf v1.5.4 h1:i7eJL8qZTpSEXOPTxNKhASYpMn+8e5Q6AdndVa1dWek=
-github.com/golang/protobuf v1.5.4/go.mod h1:lnTiLA8Wa4RWRcIUkrtSVa5nRhsEGBg48fD6rSs7xps=
-github.com/google/go-cmp v0.2.0/go.mod h1:oXzfMopK8JAjlY9xF4vHSVASa0yLyX7SntLO5aqRK0M=
-github.com/google/go-cmp v0.3.0/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
-github.com/google/go-cmp v0.3.1/go.mod h1:8QqcDgzrUqlUb/G2PQTWiueGozuR1884gddMywk6iLU=
-github.com/google/go-cmp v0.4.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
-github.com/google/go-cmp v0.5.0/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
-github.com/google/go-cmp v0.5.3/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
-github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
+github.com/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ=
 github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
 github.com/google/gofuzz v1.2.0 h1:xRy4A+RhZaiKjJ1bPfwQ8sedCA+YS2YcCHW6ec7JMi0=
-github.com/google/s2a-go v0.1.7 h1:60BLSyTrOV4/haCDW4zb1guZItoSq8foHCXrAnjBo/o=
-github.com/google/s2a-go v0.1.7/go.mod h1:50CgR4k1jNlWBu4UfS4AcfhVe1r6pdZPygJ3R8F0Qdw=
-github.com/google/uuid v1.1.2/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
 github.com/google/uuid v1.6.0 h1:NIvaJDMOsjHA8n1jAhLSgzrAzy1Hgr+hNrb57e+94F0=
 github.com/google/uuid v1.6.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
-github.com/googleapis/enterprise-certificate-proxy v0.3.2 h1:Vie5ybvEvT75RniqhfFxPRy3Bf7vr3h0cechB90XaQs=
-github.com/googleapis/enterprise-certificate-proxy v0.3.2/go.mod h1:VLSiSSBs/ksPL8kq3OBOQ6WRI2QnaFynd1DCjZ62+V0=
-github.com/googleapis/gax-go/v2 v2.12.5 h1:8gw9KZK8TiVKB6q3zHY3SBzLnrGp6HQjyfYBYGmXdxA=
-github.com/googleapis/gax-go/v2 v2.12.5/go.mod h1:BUDKcWo+RaKq5SC9vVYL0wLADa3VcfswbOMMRmB9H3E=
 github.com/gopherjs/gopherjs v1.17.2 h1:fQnZVsXk8uxXIStYb0N4bGk7jeyTalG/wsZjQ25dO0g=
 github.com/gopherjs/gopherjs v1.17.2/go.mod h1:pRRIvn/QzFLrKfvEz3qUuEhtE/zLCWfreZ6J5gM2i+k=
 github.com/gorilla/context v1.1.2 h1:WRkNAv2uoa03QNIc1A6u4O7DAGMUVoopZhkiXWA2V1o=
@ -163,8 +110,8 @@ github.com/leodido/go-urn v1.4.0 h1:WT9HwE9SGECu3lg4d/dIA+jxlljEa1/ffXKmRjqdmIQ=
 github.com/leodido/go-urn v1.4.0/go.mod h1:bvxc+MVxLKB4z00jd1z+Dvzr47oO32F/QSNjSBOlFxI=
 github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
 github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
-github.com/mattn/go-sqlite3 v1.14.22 h1:2gZY6PC6kBnID23Tichd1K+Z0oS6nE/XwU+Vz/5o4kU=
-github.com/mattn/go-sqlite3 v1.14.22/go.mod h1:Uh1q+B4BYcTPb+yiD3kU8Ct7aC0hY9fxUwlHK0RXw+Y=
+github.com/mattn/go-sqlite3 v2.0.3+incompatible h1:gXHsfypPkaMZrKbD5209QV9jbUTJKjyR5WD3HYQSd+U=
+github.com/mattn/go-sqlite3 v2.0.3+incompatible/go.mod h1:FPy6KqzDD04eiIsT53CuJW3U88zkxoIYsOqkbpncsNc=
 github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd h1:TRLaZ9cD/w8PVh93nsPXa1VrQ6jlwL5oN8l14QlcNfg=
 github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
@ -173,8 +120,6 @@ github.com/modern-go/reflect2 v1.0.2/go.mod h1:yWuevngMOJpCy52FWWMvUC8ws7m/LJsjY
 github.com/nxadm/tail v1.4.8 h1:nPr65rt6Y5JFSKQO7qToXr7pePgD6Gwiw05lkbyAQTE=
 github.com/onsi/ginkgo v1.16.5 h1:8xi0RTUf59SOSfEtZMvwTvXYMzG4gV23XVHOZiXNtnE=
 github.com/onsi/gomega v1.18.1 h1:M1GfJqGRrBrrGGsbxzV5dqM2U2ApXefZCQpkukxYRLE=
-github.com/patrickmn/go-cache v2.1.0+incompatible h1:HRMgzkcYKYpi3C8ajMPV8OFXaaRUnok+kx1WdO15EQc=
-github.com/patrickmn/go-cache v2.1.0+incompatible/go.mod h1:3Qf8kWWT7OJRJbdiICTKqZju1ZixQ/KpMGzzAfe6+WQ=
 github.com/pelletier/go-toml/v2 v2.2.2 h1:aYUidT7k73Pcl9nb2gScu7NSrKCSHIDE89b3+6Wq+LM=
 github.com/pelletier/go-toml/v2 v2.2.2/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs=
 github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
@ -183,7 +128,6 @@ github.com/pkoukk/tiktoken-go v0.1.7 h1:qOBHXX4PHtvIvmOtyg1EeKlwFRiMKAcoMp4Q+bLQ
 github.com/pkoukk/tiktoken-go v0.1.7/go.mod h1:9NiV+i9mJKGj1rYOT+njbv+ZwA/zJxYdewGl6qVatpg=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
-github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
 github.com/rogpeppe/go-internal v1.8.0 h1:FCbCCtXNOY3UtUuHUYaghJg4y7Fd14rXifAYUAtL9R8=
 github.com/smarty/assertions v1.15.0 h1:cR//PqUBUiQRakZWqBiFFQ9wb8emQGDb0HeGdqGByCY=
 github.com/smarty/assertions v1.15.0/go.mod h1:yABtdzeQs6l1brC900WlRNwj6ZR55d7B+E8C6HtKdec=
@ -205,96 +149,26 @@ github.com/twitchyliquid64/golang-asm v0.15.1 h1:SU5vSMR7hnwNxj24w34ZyCi/FmDZTkS
 github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2+aY1QWCk3Cedj/Gdt08=
 github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65EE=
 github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
-go.opencensus.io v0.24.0 h1:y73uSU6J157QMP2kn2r30vwW1A2W2WFwSCGnAVxeaD0=
-go.opencensus.io v0.24.0/go.mod h1:vNK8G9p7aAivkbmorf4v+7Hgx+Zs0yY+0fOtgBfjQKo=
-go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.49.0 h1:4Pp6oUg3+e/6M4C0A/3kJ2VYa++dsWVTtGgLVj5xtHg=
-go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.49.0/go.mod h1:Mjt1i1INqiaoZOMGR1RIUJN+i3ChKoFRqzrRQhlkbs0=
-go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.49.0 h1:jq9TW8u3so/bN+JPT166wjOI6/vQPF6Xe7nMNIltagk=
-go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.49.0/go.mod h1:p8pYQP+m5XfbZm9fxtSKAbM6oIllS7s2AfxrChvc7iw=
-go.opentelemetry.io/otel v1.24.0 h1:0LAOdjNmQeSTzGBzduGe/rU4tZhMwL5rWgtp9Ku5Jfo=
-go.opentelemetry.io/otel v1.24.0/go.mod h1:W7b9Ozg4nkF5tWI5zsXkaKKDjdVjpD4oAt9Qi/MArHo=
-go.opentelemetry.io/otel/metric v1.24.0 h1:6EhoGWWK28x1fbpA4tYTOWBkPefTDQnb8WSGXlc88kI=
-go.opentelemetry.io/otel/metric v1.24.0/go.mod h1:VYhLe1rFfxuTXLgj4CBiyz+9WYBA8pNGJgDcSFRKBco=
-go.opentelemetry.io/otel/trace v1.24.0 h1:CsKnnL4dUAr/0llH9FKuc698G04IrpWV0MQA/Y1YELI=
-go.opentelemetry.io/otel/trace v1.24.0/go.mod h1:HPc3Xr/cOApsBI154IU0OI0HJexz+aw5uPdbs3UCjNU=
 golang.org/x/arch v0.0.0-20210923205945-b76863e36670/go.mod h1:5om86z9Hs0C8fWVUuoMHwpExlXzs5Tkyp9hOrfG7pp8=
 golang.org/x/arch v0.8.0 h1:3wRIsP3pM4yUptoR96otTUOXI367OS0+c9eeRi9doIc=
 golang.org/x/arch v0.8.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys=
-golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
-golang.org/x/crypto v0.0.0-20200622213623-75b288015ac9/go.mod h1:LzIPMQfyMNhhGPhUkYOs5KpL4U8rLKemX1yGLhDgUto=
-golang.org/x/crypto v0.24.0 h1:mnl8DM0o513X8fdIkmyFE/5hTYxbwYOjDS/+rK6qpRI=
-golang.org/x/crypto v0.24.0/go.mod h1:Z1PMYSOR5nyMcyAVAIQSKCDwalqy85Aqn1x3Ws4L5DM=
-golang.org/x/exp v0.0.0-20190121172915-509febef88a4/go.mod h1:CJ0aWSM057203Lf6IL+f9T1iT9GByDxfZKAQTCR3kQA=
-golang.org/x/image v0.18.0 h1:jGzIakQa/ZXI1I0Fxvaa9W7yP25TqT6cHIHn+6CqvSQ=
-golang.org/x/image v0.18.0/go.mod h1:4yyo5vMFQjVjUcVk4jEQcU9MGy/rulF5WvUILseCM2E=
-golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
-golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU=
-golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
-golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
-golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
-golang.org/x/net v0.0.0-20190213061140-3a22650c66bd/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
-golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
-golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
-golang.org/x/net v0.0.0-20201110031124-69a78807bb2b/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
-golang.org/x/net v0.26.0 h1:soB7SVo0PWrY4vPW/+ay0jKDNScG2X9wFeYlXIvJsOQ=
-golang.org/x/net v0.26.0/go.mod h1:5YKkiSynbBIh3p6iOc/vibscux0x38BZDkn8sCUPxHE=
-golang.org/x/oauth2 v0.0.0-20180821212333-d2e6202438be/go.mod h1:N/0e6XlmueqKjAGxoOufVs8QHGRruUQn6yWY3a++T0U=
-golang.org/x/oauth2 v0.21.0 h1:tsimM75w1tF/uws5rbeHzIWxEqElMehnc+iW793zsZs=
-golang.org/x/oauth2 v0.21.0/go.mod h1:XYTD2NtWslqkgxebSiOHnXEap4TF09sJSc7H1sXbhtI=
-golang.org/x/sync v0.0.0-20180314180146-1d60e4601c6f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.0.0-20181108010431-42b317875d0f/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
+golang.org/x/crypto v0.23.0 h1:dIJU/v2J8Mdglj/8rJ6UUOM3Zc9zLZxVZwwxMooUSAI=
+golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8=
+golang.org/x/image v0.16.0 h1:9kloLAKhUufZhA12l5fwnx2NZW39/we1UhBesW433jw=
+golang.org/x/image v0.16.0/go.mod h1:ugSZItdV4nOxyqp56HmXwH0Ry0nBCpjnZdpDaIHdoPs=
+golang.org/x/net v0.25.0 h1:d/OCCoBEUq33pjydKrGQhw7IlUPI2Oylr+8qLx49kac=
+golang.org/x/net v0.25.0/go.mod h1:JkAGAh7GEvH74S6FOH42FLoXpXbE/aqXSrIQjXgsiwM=
 golang.org/x/sync v0.7.0 h1:YsImfSBoP9QPYL0xyKJPq0gcaJdG3rInoqxTWbfQu9M=
 golang.org/x/sync v0.7.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
-golang.org/x/sys v0.0.0-20180830151530-49385e6e1522/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
-golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
-golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
-golang.org/x/sys v0.0.0-20200930185726-fdedc70b468f/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.21.0 h1:rF+pYz3DAGSQAxAu1CbC7catZg4ebC4UIeIhKxBZvws=
-golang.org/x/sys v0.21.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
-golang.org/x/text v0.3.3/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
-golang.org/x/text v0.16.0 h1:a94ExnEXNtEwYLGJSIUxnWoxoRz/ZcCsV63ROupILh4=
-golang.org/x/text v0.16.0/go.mod h1:GhwF1Be+LQoKShO3cGOHzqOgRrGaYc9AvblQOmPVHnI=
-golang.org/x/time v0.5.0 h1:o7cqy6amK/52YcAKIPlM3a+Fpj35zvRj2TP+e1xFSfk=
-golang.org/x/time v0.5.0/go.mod h1:3BpzKBy/shNhVucY/MWOyx10tF3SFh9QdLuxbVysPQM=
-golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/tools v0.0.0-20190114222345-bf090417da8b/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/tools v0.0.0-20190226205152-f727befe758c/go.mod h1:9Yl7xja0Znq3iFh3HoIrodX9oNMXvdceNzlUR8zjMvY=
-golang.org/x/tools v0.0.0-20190311212946-11955173bddd/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
-golang.org/x/tools v0.0.0-20190524140312-2c0ae7006135/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
-golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
-google.golang.org/api v0.187.0 h1:Mxs7VATVC2v7CY+7Xwm4ndkX71hpElcvx0D1Ji/p1eo=
-google.golang.org/api v0.187.0/go.mod h1:KIHlTc4x7N7gKKuVsdmfBXN13yEEWXWFURWY6SBp2gk=
-google.golang.org/appengine v1.1.0/go.mod h1:EbEs0AVv82hx2wNQdGPgUI5lhzA/G0D9YwlJXL52JkM=
-google.golang.org/appengine v1.4.0/go.mod h1:xpcJRLb0r/rnEns0DIKYYv+WjYCduHsrkT7/EB5XEv4=
-google.golang.org/genproto v0.0.0-20180817151627-c66870c02cf8/go.mod h1:JiN7NxoALGmiZfu7CAH4rXhgtRTLTxftemlI0sWmxmc=
-google.golang.org/genproto v0.0.0-20190819201941-24fa4b261c55/go.mod h1:DMBHOl98Agz4BDEuKkezgsaosCRResVns1a3J2ZsMNc=
-google.golang.org/genproto v0.0.0-20200526211855-cb27e3aa2013/go.mod h1:NbSheEEYHJ7i3ixzK3sjbqSGDJWnxyFXZblF3eUsNvo=
-google.golang.org/genproto/googleapis/api v0.0.0-20240617180043-68d350f18fd4 h1:MuYw1wJzT+ZkybKfaOXKp5hJiZDn2iHaXRw0mRYdHSc=
-google.golang.org/genproto/googleapis/api v0.0.0-20240617180043-68d350f18fd4/go.mod h1:px9SlOOZBg1wM1zdnr8jEL4CNGUBZ+ZKYtNPApNQc4c=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20240624140628-dc46fd24d27d h1:k3zyW3BYYR30e8v3x0bTDdE9vpYFjZHK+HcyqkrppWk=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20240624140628-dc46fd24d27d/go.mod h1:Ue6ibwXGpU+dqIcODieyLOcgj7z8+IcskoNIgZxtrFY=
-google.golang.org/grpc v1.19.0/go.mod h1:mqu4LbDTu4XGKhr4mRzUsmM4RtVoemTSY81AxZiDr8c=
-google.golang.org/grpc v1.23.0/go.mod h1:Y5yQAOtifL1yxbo5wqy6BxZv8vAUGQwXBOALyacEbxg=
-google.golang.org/grpc v1.25.1/go.mod h1:c3i+UQWmh7LiEpx4sFZnkU36qjEYZ0imhYfXVyQciAY=
-google.golang.org/grpc v1.27.0/go.mod h1:qbnxyOmOxrQa7FizSgH+ReBfzJrCY1pSN7KXBS8abTk=
-google.golang.org/grpc v1.33.2/go.mod h1:JMHMWHQWaTccqQQlmk3MJZS+GWXOdAesneDmEnv2fbc=
-google.golang.org/grpc v1.64.1 h1:LKtvyfbX3UGVPFcGqJ9ItpVWW6oN/2XqTxfAnwRRXiA=
-google.golang.org/grpc v1.64.1/go.mod h1:hiQF4LFZelK2WKaP6W0L92zGHtiQdZxk8CrSdvyjeP0=
-google.golang.org/protobuf v0.0.0-20200109180630-ec00e32a8dfd/go.mod h1:DFci5gLYBciE7Vtevhsrf46CRTquxDuWsQurQQe4oz8=
-google.golang.org/protobuf v0.0.0-20200221191635-4d8936d0db64/go.mod h1:kwYJMbMJ01Woi6D6+Kah6886xMZcty6N08ah7+eCXa0=
-google.golang.org/protobuf v0.0.0-20200228230310-ab0ca4ff8a60/go.mod h1:cfTl7dwQJ+fmap5saPgwCLgHXTUD7jkjRqWcaiX5VyM=
-google.golang.org/protobuf v1.20.1-0.20200309200217-e05f789c0967/go.mod h1:A+miEFZTKqfCUM6K7xSMQL9OKL/b6hQv+e19PK+JZNE=
-google.golang.org/protobuf v1.21.0/go.mod h1:47Nbq4nVaFHyn7ilMalzfO3qCViNmqZ2kzikPIcrTAo=
-google.golang.org/protobuf v1.22.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
-google.golang.org/protobuf v1.23.0/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
-google.golang.org/protobuf v1.23.1-0.20200526195155-81db48ad09cc/go.mod h1:EGpADcykh3NcUnDUJcl1+ZksZNG86OlYog2l/sGQquU=
-google.golang.org/protobuf v1.25.0/go.mod h1:9JNX74DMeImyA3h4bdi1ymwjUzf21/xIlbajtzgsN7c=
-google.golang.org/protobuf v1.34.2 h1:6xV6lTsCfpGD21XK49h7MhtcApnLqkfYgPcdHftf6hg=
-google.golang.org/protobuf v1.34.2/go.mod h1:qYOHts0dSfpeUzUFpOMr/WGzszTmLH+DiWniOlNbLDw=
+golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y=
+golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/text v0.15.0 h1:h1V/4gjBv8v9cjcR6+AR5+/cIYK5N/WAgiv4xlsEtAk=
+golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
+golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE=
+google.golang.org/protobuf v1.34.1 h1:9ddQBjfCyZPOHPUiPxpYESBLc+T8P3E+Vo4IbKZgFWg=
+google.golang.org/protobuf v1.34.1/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/check.v1 v1.0.0-20201130134442-10cb98267c6c h1:Hei/4ADfdWqJk1ZMxUNpqntNwaWcugrBjAiHlqqRiVk=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 h1:uRGJdciOHaEIrze2W8Q3AKkepLTh2hOroT7a+7czfdQ=
@ -311,7 +185,5 @@ gorm.io/driver/sqlite v1.5.5/go.mod h1:6NgQ7sQWAIFsPrJJl1lSNSu2TABh0ZZ/zm5fosATa
 gorm.io/gorm v1.25.7/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8=
 gorm.io/gorm v1.25.10 h1:dQpO+33KalOA+aFYGlK+EfxcI5MbO7EP2yYygwh9h+s=
 gorm.io/gorm v1.25.10/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8=
-honnef.co/go/tools v0.0.0-20190102054323-c2f93a96b099/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
-honnef.co/go/tools v0.0.0-20190523083050-ea95bdfd59fc/go.mod h1:rf3lG4BRIbNafJWhAfAdb/ePZxsR/4RtNHQocxwk9r4=
 nullprogram.com/x/optparse v1.0.0/go.mod h1:KdyPE+Igbe0jQUrVfMqDMeJQIJZEuyV7pjYmp6pbG50=
 rsc.io/pdf v0.1.1/go.mod h1:n8OzWcQ6Sp37PL01nO98y4iUCRdTGarVfzxY20ICaU4=
--- a/main.go
+++ b/main.go
@ -27,19 +27,27 @@ func main() {
 	common.Init()
 	logger.SetupLogger()
 	logger.SysLogf("One API %s started", common.Version)
-
-	if os.Getenv("GIN_MODE") != gin.DebugMode {
+	if os.Getenv("GIN_MODE") != "debug" {
 		gin.SetMode(gin.ReleaseMode)
 	}
 	if config.DebugEnabled {
 		logger.SysLog("running in debug mode")
 	}
-
-	// Initialize SQL Database
-	model.InitDB()
-	model.InitLogDB()
-
 	var err error
+	// Initialize SQL Database
+	model.DB, err = model.InitDB("SQL_DSN")
+	if err != nil {
+		logger.FatalLog("failed to initialize database: " + err.Error())
+	}
+	if os.Getenv("LOG_SQL_DSN") != "" {
+		logger.SysLog("using secondary database for table logs")
+		model.LOG_DB, err = model.InitDB("LOG_SQL_DSN")
+		if err != nil {
+			logger.FatalLog("failed to initialize secondary database: " + err.Error())
+		}
+	} else {
+		model.LOG_DB = model.DB
+	}
 	err = model.CreateRootAccountIfNeed()
 	if err != nil {
 		logger.FatalLog("database init error: " + err.Error())
--- a/middleware/auth.go
+++ b/middleware/auth.go
@ -140,12 +140,6 @@ func TokenAuth() func(c *gin.Context) {
 				return
 			}
 		}
-
-		// set channel id for proxy relay
-		if channelId := c.Param("channelid"); channelId != "" {
-			c.Set(ctxkey.SpecificChannelId, channelId)
-		}
-
 		c.Next()
 	}
 }
--- a/middleware/distributor.go
+++ b/middleware/distributor.go
@ -12,7 +12,7 @@ import (
 )

 type ModelRequest struct {
-	Model string `json:"model" form:"model"`
+	Model string `json:"model"`
 }

 func Distribute() func(c *gin.Context) {
--- a/middleware/rate-limit.go
+++ b/middleware/rate-limit.go
@ -3,12 +3,11 @@ package middleware
 import (
 	"context"
 	"fmt"
-	"net/http"
-	"time"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
+	"net/http"
+	"time"
 )

 var timeFormat = "2006-01-02T15:04:05.000Z"
@ -71,11 +70,6 @@ func memoryRateLimiter(c *gin.Context, maxRequestNum int, duration int64, mark s
 }

 func rateLimitFactory(maxRequestNum int, duration int64, mark string) func(c *gin.Context) {
-	if maxRequestNum == 0 {
-		return func(c *gin.Context) {
-			c.Next()
-		}
-	}
 	if common.RedisEnabled {
 		return func(c *gin.Context) {
 			redisRateLimiter(c, maxRequestNum, duration, mark)
--- a/model/channel.go
+++ b/model/channel.go
@ -3,7 +3,6 @@ package model
 import (
 	"encoding/json"
 	"fmt"
-
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
@ -40,15 +39,13 @@ type Channel struct {
 }

 type ChannelConfig struct {
-	Region            string `json:"region,omitempty"`
-	SK                string `json:"sk,omitempty"`
-	AK                string `json:"ak,omitempty"`
-	UserID            string `json:"user_id,omitempty"`
-	APIVersion        string `json:"api_version,omitempty"`
-	LibraryID         string `json:"library_id,omitempty"`
-	Plugin            string `json:"plugin,omitempty"`
-	VertexAIProjectID string `json:"vertex_ai_project_id,omitempty"`
-	VertexAIADC       string `json:"vertex_ai_adc,omitempty"`
+	Region     string `json:"region,omitempty"`
+	SK         string `json:"sk,omitempty"`
+	AK         string `json:"ak,omitempty"`
+	UserID     string `json:"user_id,omitempty"`
+	APIVersion string `json:"api_version,omitempty"`
+	LibraryID  string `json:"library_id,omitempty"`
+	Plugin     string `json:"plugin,omitempty"`
 }

 func GetAllChannels(startIdx int, num int, scope string) ([]*Channel, error) {
--- a/model/log.go
+++ b/model/log.go
@ -3,7 +3,6 @@ package model
 import (
 	"context"
 	"fmt"
-
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/helper"
@ -153,11 +152,7 @@ func SearchUserLogs(userId int, keyword string) (logs []*Log, err error) {
 }

 func SumUsedQuota(logType int, startTimestamp int64, endTimestamp int64, modelName string, username string, tokenName string, channel int) (quota int64) {
-	ifnull := "ifnull"
-	if common.UsingPostgreSQL {
-		ifnull = "COALESCE"
-	}
-	tx := LOG_DB.Table("logs").Select(fmt.Sprintf("%s(sum(quota),0)", ifnull))
+	tx := LOG_DB.Table("logs").Select("ifnull(sum(quota),0)")
 	if username != "" {
 		tx = tx.Where("username = ?", username)
 	}
@ -181,11 +176,7 @@ func SumUsedQuota(logType int, startTimestamp int64, endTimestamp int64, modelNa
 }

 func SumUsedToken(logType int, startTimestamp int64, endTimestamp int64, modelName string, username string, tokenName string) (token int) {
-	ifnull := "ifnull"
-	if common.UsingPostgreSQL {
-		ifnull = "COALESCE"
-	}
-	tx := LOG_DB.Table("logs").Select(fmt.Sprintf("%s(sum(prompt_tokens),0) + %s(sum(completion_tokens),0)", ifnull, ifnull))
+	tx := LOG_DB.Table("logs").Select("ifnull(sum(prompt_tokens),0) + ifnull(sum(completion_tokens),0)")
 	if username != "" {
 		tx = tx.Where("username = ?", username)
 	}
--- a/model/main.go
+++ b/model/main.go
@ -1,7 +1,6 @@
 package model

 import (
-	"database/sql"
 	"fmt"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
@ -30,17 +29,13 @@ func CreateRootAccountIfNeed() error {
 		if err != nil {
 			return err
 		}
-		accessToken := random.GetUUID()
-		if config.InitialRootAccessToken != "" {
-			accessToken = config.InitialRootAccessToken
-		}
 		rootUser := User{
 			Username:    "root",
 			Password:    hashedPassword,
 			Role:        RoleRootUser,
 			Status:      UserStatusEnabled,
 			DisplayName: "Root User",
-			AccessToken: accessToken,
+			AccessToken: random.GetUUID(),
 			Quota:       500000000000000,
 		}
 		DB.Create(&rootUser)
@ -65,156 +60,90 @@ func CreateRootAccountIfNeed() error {
 }

 func chooseDB(envName string) (*gorm.DB, error) {
-	dsn := os.Getenv(envName)
-
-	switch {
-	case strings.HasPrefix(dsn, "postgres://"):
-		// Use PostgreSQL
-		return openPostgreSQL(dsn)
-	case dsn != "":
+	if os.Getenv(envName) != "" {
+		dsn := os.Getenv(envName)
+		if strings.HasPrefix(dsn, "postgres://") {
+			// Use PostgreSQL
+			logger.SysLog("using PostgreSQL as database")
+			common.UsingPostgreSQL = true
+			return gorm.Open(postgres.New(postgres.Config{
+				DSN:                  dsn,
+				PreferSimpleProtocol: true, // disables implicit prepared statement usage
+			}), &gorm.Config{
+				PrepareStmt: true, // precompile SQL
+			})
+		}
 		// Use MySQL
-		return openMySQL(dsn)
-	default:
-		// Use SQLite
-		return openSQLite()
+		logger.SysLog("using MySQL as database")
+		common.UsingMySQL = true
+		return gorm.Open(mysql.Open(dsn), &gorm.Config{
+			PrepareStmt: true, // precompile SQL
+		})
 	}
-}
-
-func openPostgreSQL(dsn string) (*gorm.DB, error) {
-	logger.SysLog("using PostgreSQL as database")
-	common.UsingPostgreSQL = true
-	return gorm.Open(postgres.New(postgres.Config{
-		DSN:                  dsn,
-		PreferSimpleProtocol: true, // disables implicit prepared statement usage
-	}), &gorm.Config{
-		PrepareStmt: true, // precompile SQL
-	})
-}
-
-func openMySQL(dsn string) (*gorm.DB, error) {
-	logger.SysLog("using MySQL as database")
-	common.UsingMySQL = true
-	return gorm.Open(mysql.Open(dsn), &gorm.Config{
-		PrepareStmt: true, // precompile SQL
-	})
-}
-
-func openSQLite() (*gorm.DB, error) {
+	// Use SQLite
 	logger.SysLog("SQL_DSN not set, using SQLite as database")
 	common.UsingSQLite = true
-	dsn := fmt.Sprintf("%s?_busy_timeout=%d", common.SQLitePath, common.SQLiteBusyTimeout)
-	return gorm.Open(sqlite.Open(dsn), &gorm.Config{
+	config := fmt.Sprintf("?_busy_timeout=%d", common.SQLiteBusyTimeout)
+	return gorm.Open(sqlite.Open(common.SQLitePath+config), &gorm.Config{
 		PrepareStmt: true, // precompile SQL
 	})
 }

-func InitDB() {
-	var err error
-	DB, err = chooseDB("SQL_DSN")
-	if err != nil {
-		logger.FatalLog("failed to initialize database: " + err.Error())
-		return
-	}
+func InitDB(envName string) (db *gorm.DB, err error) {
+	db, err = chooseDB(envName)
+	if err == nil {
+		if config.DebugSQLEnabled {
+			db = db.Debug()
+		}
+		sqlDB, err := db.DB()
+		if err != nil {
+			return nil, err
+		}
+		sqlDB.SetMaxIdleConns(env.Int("SQL_MAX_IDLE_CONNS", 100))
+		sqlDB.SetMaxOpenConns(env.Int("SQL_MAX_OPEN_CONNS", 1000))
+		sqlDB.SetConnMaxLifetime(time.Second * time.Duration(env.Int("SQL_MAX_LIFETIME", 60)))

-	sqlDB := setDBConns(DB)
-
-	if !config.IsMasterNode {
-		return
+		if !config.IsMasterNode {
+			return db, err
+		}
+		if common.UsingMySQL {
+			_, _ = sqlDB.Exec("DROP INDEX idx_channels_key ON channels;") // TODO: delete this line when most users have upgraded
+		}
+		logger.SysLog("database migration started")
+		err = db.AutoMigrate(&Channel{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&Token{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&User{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&Option{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&Redemption{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&Ability{})
+		if err != nil {
+			return nil, err
+		}
+		err = db.AutoMigrate(&Log{})
+		if err != nil {
+			return nil, err
+		}
+		logger.SysLog("database migrated")
+		return db, err
+	} else {
+		logger.FatalLog(err)
 	}
-
-	if common.UsingMySQL {
-		_, _ = sqlDB.Exec("DROP INDEX idx_channels_key ON channels;") // TODO: delete this line when most users have upgraded
-	}
-
-	logger.SysLog("database migration started")
-	if err = migrateDB(); err != nil {
-		logger.FatalLog("failed to migrate database: " + err.Error())
-		return
-	}
-	logger.SysLog("database migrated")
-}
-
-func migrateDB() error {
-	var err error
-	if err = DB.AutoMigrate(&Channel{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Token{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&User{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Option{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Redemption{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Ability{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Log{}); err != nil {
-		return err
-	}
-	if err = DB.AutoMigrate(&Channel{}); err != nil {
-		return err
-	}
-	return nil
-}
-
-func InitLogDB() {
-	if os.Getenv("LOG_SQL_DSN") == "" {
-		LOG_DB = DB
-		return
-	}
-
-	logger.SysLog("using secondary database for table logs")
-	var err error
-	LOG_DB, err = chooseDB("LOG_SQL_DSN")
-	if err != nil {
-		logger.FatalLog("failed to initialize secondary database: " + err.Error())
-		return
-	}
-
-	setDBConns(LOG_DB)
-
-	if !config.IsMasterNode {
-		return
-	}
-
-	logger.SysLog("secondary database migration started")
-	err = migrateLOGDB()
-	if err != nil {
-		logger.FatalLog("failed to migrate secondary database: " + err.Error())
-		return
-	}
-	logger.SysLog("secondary database migrated")
-}
-
-func migrateLOGDB() error {
-	var err error
-	if err = LOG_DB.AutoMigrate(&Log{}); err != nil {
-		return err
-	}
-	return nil
-}
-
-func setDBConns(db *gorm.DB) *sql.DB {
-	if config.DebugSQLEnabled {
-		db = db.Debug()
-	}
-
-	sqlDB, err := db.DB()
-	if err != nil {
-		logger.FatalLog("failed to connect database: " + err.Error())
-		return nil
-	}
-
-	sqlDB.SetMaxIdleConns(env.Int("SQL_MAX_IDLE_CONNS", 100))
-	sqlDB.SetMaxOpenConns(env.Int("SQL_MAX_OPEN_CONNS", 1000))
-	sqlDB.SetConnMaxLifetime(time.Second * time.Duration(env.Int("SQL_MAX_LIFETIME", 60)))
-	return sqlDB
+	return db, err
 }

 func closeDB(db *gorm.DB) error {
--- a/model/option.go
+++ b/model/option.go
@ -28,7 +28,6 @@ func InitOptionMap() {
 	config.OptionMap["PasswordRegisterEnabled"] = strconv.FormatBool(config.PasswordRegisterEnabled)
 	config.OptionMap["EmailVerificationEnabled"] = strconv.FormatBool(config.EmailVerificationEnabled)
 	config.OptionMap["GitHubOAuthEnabled"] = strconv.FormatBool(config.GitHubOAuthEnabled)
-	config.OptionMap["OidcEnabled"] = strconv.FormatBool(config.OidcEnabled)
 	config.OptionMap["WeChatAuthEnabled"] = strconv.FormatBool(config.WeChatAuthEnabled)
 	config.OptionMap["TurnstileCheckEnabled"] = strconv.FormatBool(config.TurnstileCheckEnabled)
 	config.OptionMap["RegisterEnabled"] = strconv.FormatBool(config.RegisterEnabled)
@ -131,8 +130,6 @@ func updateOptionMap(key string, value string) (err error) {
 			config.EmailVerificationEnabled = boolValue
 		case "GitHubOAuthEnabled":
 			config.GitHubOAuthEnabled = boolValue
-		case "OidcEnabled":
-			config.OidcEnabled = boolValue
 		case "WeChatAuthEnabled":
 			config.WeChatAuthEnabled = boolValue
 		case "TurnstileCheckEnabled":
@ -179,18 +176,6 @@ func updateOptionMap(key string, value string) (err error) {
 		config.LarkClientId = value
 	case "LarkClientSecret":
 		config.LarkClientSecret = value
-	case "OidcClientId":
-		config.OidcClientId = value
-	case "OidcClientSecret":
-		config.OidcClientSecret = value
-	case "OidcWellKnown":
-		config.OidcWellKnown = value
-	case "OidcAuthorizationEndpoint":
-		config.OidcAuthorizationEndpoint = value
-	case "OidcTokenEndpoint":
-		config.OidcTokenEndpoint = value
-	case "OidcUserinfoEndpoint":
-		config.OidcUserinfoEndpoint = value
 	case "Footer":
 		config.Footer = value
 	case "SystemName":
--- a/model/token.go
+++ b/model/token.go
@ -30,7 +30,7 @@ type Token struct {
 	RemainQuota    int64   `json:"remain_quota" gorm:"bigint;default:0"`
 	UnlimitedQuota bool    `json:"unlimited_quota" gorm:"default:false"`
 	UsedQuota      int64   `json:"used_quota" gorm:"bigint;default:0"` // used quota
-	Models         *string `json:"models" gorm:"type:text"`            // allowed models
+	Models         *string `json:"models" gorm:"default:''"`           // allowed models
 	Subnet         *string `json:"subnet" gorm:"default:''"`           // allowed subnet
 }

@ -121,40 +121,30 @@ func GetTokenById(id int) (*Token, error) {
 	return &token, err
 }

-func (t *Token) Insert() error {
+func (token *Token) Insert() error {
 	var err error
-	err = DB.Create(t).Error
+	err = DB.Create(token).Error
 	return err
 }

 // Update Make sure your token's fields is completed, because this will update non-zero values
-func (t *Token) Update() error {
+func (token *Token) Update() error {
 	var err error
-	err = DB.Model(t).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota", "models", "subnet").Updates(t).Error
+	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota", "models", "subnet").Updates(token).Error
 	return err
 }

-func (t *Token) SelectUpdate() error {
+func (token *Token) SelectUpdate() error {
 	// This can update zero values
-	return DB.Model(t).Select("accessed_time", "status").Updates(t).Error
+	return DB.Model(token).Select("accessed_time", "status").Updates(token).Error
 }

-func (t *Token) Delete() error {
+func (token *Token) Delete() error {
 	var err error
-	err = DB.Delete(t).Error
+	err = DB.Delete(token).Error
 	return err
 }

-func (t *Token) GetModels() string {
-	if t == nil {
-		return ""
-	}
-	if t.Models == nil {
-		return ""
-	}
-	return *t.Models
-}
-
 func DeleteTokenById(id int, userId int) (err error) {
 	// Why we need userId here? In case user want to delete other's token.
 	if id == 0 || userId == 0 {
@ -264,14 +254,14 @@ func PreConsumeTokenQuota(tokenId int, quota int64) (err error) {

 func PostConsumeTokenQuota(tokenId int, quota int64) (err error) {
 	token, err := GetTokenById(tokenId)
-	if err != nil {
-		return err
-	}
 	if quota > 0 {
 		err = DecreaseUserQuota(token.UserId, quota)
 	} else {
 		err = IncreaseUserQuota(token.UserId, -quota)
 	}
+	if err != nil {
+		return err
+	}
 	if !token.UnlimitedQuota {
 		if quota > 0 {
 			err = DecreaseTokenQuota(tokenId, quota)
--- a/model/user.go
+++ b/model/user.go
@ -39,7 +39,6 @@ type User struct {
 	GitHubId         string `json:"github_id" gorm:"column:github_id;index"`
 	WeChatId         string `json:"wechat_id" gorm:"column:wechat_id;index"`
 	LarkId           string `json:"lark_id" gorm:"column:lark_id;index"`
-	OidcId           string `json:"oidc_id" gorm:"column:oidc_id;index"`
 	VerificationCode string `json:"verification_code" gorm:"-:all"`                                    // this field is only for Email verification, don't save it to database!
 	AccessToken      string `json:"access_token" gorm:"type:char(32);column:access_token;uniqueIndex"` // this token is for system management
 	Quota            int64  `json:"quota" gorm:"bigint;default:0"`
@ -246,14 +245,6 @@ func (user *User) FillUserByLarkId() error {
 	return nil
 }

-func (user *User) FillUserByOidcId() error {
-	if user.OidcId == "" {
-		return errors.New("oidc id 为空！")
-	}
-	DB.Where(User{OidcId: user.OidcId}).First(user)
-	return nil
-}
-
 func (user *User) FillUserByWeChatId() error {
 	if user.WeChatId == "" {
 		return errors.New("WeChat id 为空！")
@ -286,10 +277,6 @@ func IsLarkIdAlreadyTaken(githubId string) bool {
 	return DB.Where("lark_id = ?", githubId).Find(&User{}).RowsAffected == 1
 }

-func IsOidcIdAlreadyTaken(oidcId string) bool {
-	return DB.Where("oidc_id = ?", oidcId).Find(&User{}).RowsAffected == 1
-}
-
 func IsUsernameAlreadyTaken(username string) bool {
 	return DB.Where("username = ?", username).Find(&User{}).RowsAffected == 1
 }
--- a/monitor/manage.go
+++ b/monitor/manage.go
@ -1,11 +1,10 @@
 package monitor

 import (
-	"net/http"
-	"strings"
-
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/relay/model"
+	"net/http"
+	"strings"
 )

 func ShouldDisableChannel(err *model.Error, statusCode int) bool {
@ -19,23 +18,31 @@ func ShouldDisableChannel(err *model.Error, statusCode int) bool {
 		return true
 	}
 	switch err.Type {
-	case "insufficient_quota", "authentication_error", "permission_error", "forbidden":
+	case "insufficient_quota":
+		return true
+	// https://docs.anthropic.com/claude/reference/errors
+	case "authentication_error":
+		return true
+	case "permission_error":
+		return true
+	case "forbidden":
 		return true
 	}
 	if err.Code == "invalid_api_key" || err.Code == "account_deactivated" {
 		return true
 	}
-
-	lowerMessage := strings.ToLower(err.Message)
-	if strings.Contains(lowerMessage, "your access was terminated") ||
-		strings.Contains(lowerMessage, "violation of our policies") ||
-		strings.Contains(lowerMessage, "your credit balance is too low") ||
-		strings.Contains(lowerMessage, "organization has been disabled") ||
-		strings.Contains(lowerMessage, "credit") ||
-		strings.Contains(lowerMessage, "balance") ||
-		strings.Contains(lowerMessage, "permission denied") ||
-  	strings.Contains(lowerMessage, "organization has been restricted") || // groq
-		strings.Contains(lowerMessage, "已欠费") {
+	if strings.HasPrefix(err.Message, "Your credit balance is too low") { // anthropic
+		return true
+	} else if strings.HasPrefix(err.Message, "This organization has been disabled.") {
+		return true
+	}
+	//if strings.Contains(err.Message, "quota") {
+	//	return true
+	//}
+	if strings.Contains(err.Message, "credit") {
+		return true
+	}
+	if strings.Contains(err.Message, "balance") {
 		return true
 	}
 	return false
--- a/relay/adaptor.go
+++ b/relay/adaptor.go
@ -15,9 +15,7 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/ollama"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/adaptor/palm"
-	"github.com/songquanpeng/one-api/relay/adaptor/proxy"
 	"github.com/songquanpeng/one-api/relay/adaptor/tencent"
-	"github.com/songquanpeng/one-api/relay/adaptor/vertexai"
 	"github.com/songquanpeng/one-api/relay/adaptor/xunfei"
 	"github.com/songquanpeng/one-api/relay/adaptor/zhipu"
 	"github.com/songquanpeng/one-api/relay/apitype"
@ -57,10 +55,6 @@ func GetAdaptor(apiType int) adaptor.Adaptor {
 		return &cloudflare.Adaptor{}
 	case apitype.DeepL:
 		return &deepl.Adaptor{}
-	case apitype.VertexAI:
-		return &vertexai.Adaptor{}
-	case apitype.Proxy:
-		return &proxy.Adaptor{}
 	}
 	return nil
 }
--- a/relay/adaptor/aiproxy/main.go
+++ b/relay/adaptor/aiproxy/main.go
@ -4,12 +4,6 @@ import (
 	"bufio"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strconv"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
@ -18,6 +12,10 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
 )

 // https://docs.aiproxy.io/dev/library#使用已经定制好的知识库进行对话问答
@ -91,7 +89,6 @@ func streamResponseAIProxyLibrary2OpenAI(response *LibraryStreamResponse) *opena

 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	var usage model.Usage
-	var documents []LibraryDocument
 	scanner := bufio.NewScanner(resp.Body)
 	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		if atEOF && len(data) == 0 {
@ -105,48 +102,60 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 5 { // ignore blank line or wrong format
+				continue
+			}
+			if data[:5] != "data:" {
+				continue
+			}
+			data = data[5:]
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 5 || data[:5] != "data:" {
-			continue
+	var documents []LibraryDocument
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var AIProxyLibraryResponse LibraryStreamResponse
+			err := json.Unmarshal([]byte(data), &AIProxyLibraryResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			if len(AIProxyLibraryResponse.Documents) != 0 {
+				documents = AIProxyLibraryResponse.Documents
+			}
+			response := streamResponseAIProxyLibrary2OpenAI(&AIProxyLibraryResponse)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			response := documentsAIProxyLibrary(documents)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = data[5:]
-
-		var AIProxyLibraryResponse LibraryStreamResponse
-		err := json.Unmarshal([]byte(data), &AIProxyLibraryResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-		if len(AIProxyLibraryResponse.Documents) != 0 {
-			documents = AIProxyLibraryResponse.Documents
-		}
-		response := streamResponseAIProxyLibrary2OpenAI(&AIProxyLibraryResponse)
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	response := documentsAIProxyLibrary(documents)
-	err := render.ObjectData(c, response)
-	if err != nil {
-		logger.SysError(err.Error())
-	}
-	render.Done(c)
-
-	err = resp.Body.Close()
+	})
+	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
-
 	return nil, &usage
 }

--- a/relay/adaptor/ali/main.go
+++ b/relay/adaptor/ali/main.go
@ -3,18 +3,15 @@ package ali
 import (
 	"bufio"
 	"encoding/json"
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
 )

 // https://help.aliyun.com/document_detail/613695.html?spm=a2c4g.2399480.0.0.1adb778fAdzP9w#341800c0f8w0r
@ -60,7 +57,7 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {

 func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
 	return &EmbeddingRequest{
-		Model: request.Model,
+		Model: "text-embedding-v1",
 		Input: struct {
 			Texts []string `json:"texts"`
 		}{
@ -103,9 +100,8 @@ func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStat
 			StatusCode: resp.StatusCode,
 		}, nil
 	}
-	requestModel := c.GetString(ctxkey.RequestModel)
+
 	fullTextResponse := embeddingResponseAli2OpenAI(&aliResponse)
-	fullTextResponse.Model = requestModel
 	jsonResponse, err := json.Marshal(fullTextResponse)
 	if err != nil {
 		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
@ -185,43 +181,56 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 5 { // ignore blank line or wrong format
+				continue
+			}
+			if data[:5] != "data:" {
+				continue
+			}
+			data = data[5:]
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 5 || data[:5] != "data:" {
-			continue
+	//lastResponseText := ""
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var aliResponse ChatResponse
+			err := json.Unmarshal([]byte(data), &aliResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			if aliResponse.Usage.OutputTokens != 0 {
+				usage.PromptTokens = aliResponse.Usage.InputTokens
+				usage.CompletionTokens = aliResponse.Usage.OutputTokens
+				usage.TotalTokens = aliResponse.Usage.InputTokens + aliResponse.Usage.OutputTokens
+			}
+			response := streamResponseAli2OpenAI(&aliResponse)
+			if response == nil {
+				return true
+			}
+			//response.Choices[0].Delta.Content = strings.TrimPrefix(response.Choices[0].Delta.Content, lastResponseText)
+			//lastResponseText = aliResponse.Output.Text
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = data[5:]
-
-		var aliResponse ChatResponse
-		err := json.Unmarshal([]byte(data), &aliResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-		if aliResponse.Usage.OutputTokens != 0 {
-			usage.PromptTokens = aliResponse.Usage.InputTokens
-			usage.CompletionTokens = aliResponse.Usage.OutputTokens
-			usage.TotalTokens = aliResponse.Usage.InputTokens + aliResponse.Usage.OutputTokens
-		}
-		response := streamResponseAli2OpenAI(&aliResponse)
-		if response == nil {
-			continue
-		}
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
--- a/relay/adaptor/anthropic/adaptor.go
+++ b/relay/adaptor/anthropic/adaptor.go
@ -3,14 +3,12 @@ package anthropic
 import (
 	"errors"
 	"fmt"
-	"io"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
 )

 type Adaptor struct {
@ -33,13 +31,6 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *me
 	}
 	req.Header.Set("anthropic-version", anthropicVersion)
 	req.Header.Set("anthropic-beta", "messages-2023-12-15")
-
-	// https://x.com/alexalbert__/status/1812921642143900036
-	// claude-3-5-sonnet can support 8k context
-	if strings.HasPrefix(meta.ActualModelName, "claude-3-5-sonnet") {
-		req.Header.Set("anthropic-beta", "max-tokens-3-5-sonnet-2024-07-15")
-	}
-
 	return nil
 }

--- a/relay/adaptor/anthropic/constants.go
+++ b/relay/adaptor/anthropic/constants.go
@ -5,5 +5,4 @@ var ModelList = []string{
 	"claude-3-haiku-20240307",
 	"claude-3-sonnet-20240229",
 	"claude-3-opus-20240229",
-	"claude-3-5-sonnet-20240620",
 }
--- a/relay/adaptor/anthropic/main.go
+++ b/relay/adaptor/anthropic/main.go
@ -4,7 +4,6 @@ import (
 	"bufio"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
 	"io"
 	"net/http"
 	"strings"
@ -29,30 +28,12 @@ func stopReasonClaude2OpenAI(reason *string) string {
 		return "stop"
 	case "max_tokens":
 		return "length"
-	case "tool_use":
-		return "tool_calls"
 	default:
 		return *reason
 	}
 }

 func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
-	claudeTools := make([]Tool, 0, len(textRequest.Tools))
-
-	for _, tool := range textRequest.Tools {
-		if params, ok := tool.Function.Parameters.(map[string]any); ok {
-			claudeTools = append(claudeTools, Tool{
-				Name:        tool.Function.Name,
-				Description: tool.Function.Description,
-				InputSchema: InputSchema{
-					Type:       params["type"].(string),
-					Properties: params["properties"],
-					Required:   params["required"],
-				},
-			})
-		}
-	}
-
 	claudeRequest := Request{
 		Model:       textRequest.Model,
 		MaxTokens:   textRequest.MaxTokens,
@ -60,24 +41,6 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
 		TopP:        textRequest.TopP,
 		TopK:        textRequest.TopK,
 		Stream:      textRequest.Stream,
-		Tools:       claudeTools,
-	}
-	if len(claudeTools) > 0 {
-		claudeToolChoice := struct {
-			Type string `json:"type"`
-			Name string `json:"name,omitempty"`
-		}{Type: "auto"} // default value https://docs.anthropic.com/en/docs/build-with-claude/tool-use#controlling-claudes-output
-		if choice, ok := textRequest.ToolChoice.(map[string]any); ok {
-			if function, ok := choice["function"].(map[string]any); ok {
-				claudeToolChoice.Type = "tool"
-				claudeToolChoice.Name = function["name"].(string)
-			}
-		} else if toolChoiceType, ok := textRequest.ToolChoice.(string); ok {
-			if toolChoiceType == "any" {
-				claudeToolChoice.Type = toolChoiceType
-			}
-		}
-		claudeRequest.ToolChoice = claudeToolChoice
 	}
 	if claudeRequest.MaxTokens == 0 {
 		claudeRequest.MaxTokens = 4096
@ -100,24 +63,7 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
 		if message.IsStringContent() {
 			content.Type = "text"
 			content.Text = message.StringContent()
-			if message.Role == "tool" {
-				claudeMessage.Role = "user"
-				content.Type = "tool_result"
-				content.Content = content.Text
-				content.Text = ""
-				content.ToolUseId = message.ToolCallId
-			}
 			claudeMessage.Content = append(claudeMessage.Content, content)
-			for i := range message.ToolCalls {
-				inputParam := make(map[string]any)
-				_ = json.Unmarshal([]byte(message.ToolCalls[i].Function.Arguments.(string)), &inputParam)
-				claudeMessage.Content = append(claudeMessage.Content, Content{
-					Type:  "tool_use",
-					Id:    message.ToolCalls[i].Id,
-					Name:  message.ToolCalls[i].Function.Name,
-					Input: inputParam,
-				})
-			}
 			claudeRequest.Messages = append(claudeRequest.Messages, claudeMessage)
 			continue
 		}
@ -150,35 +96,16 @@ func StreamResponseClaude2OpenAI(claudeResponse *StreamResponse) (*openai.ChatCo
 	var response *Response
 	var responseText string
 	var stopReason string
-	tools := make([]model.Tool, 0)
-
 	switch claudeResponse.Type {
 	case "message_start":
 		return nil, claudeResponse.Message
 	case "content_block_start":
 		if claudeResponse.ContentBlock != nil {
 			responseText = claudeResponse.ContentBlock.Text
-			if claudeResponse.ContentBlock.Type == "tool_use" {
-				tools = append(tools, model.Tool{
-					Id:   claudeResponse.ContentBlock.Id,
-					Type: "function",
-					Function: model.Function{
-						Name:      claudeResponse.ContentBlock.Name,
-						Arguments: "",
-					},
-				})
-			}
 		}
 	case "content_block_delta":
 		if claudeResponse.Delta != nil {
 			responseText = claudeResponse.Delta.Text
-			if claudeResponse.Delta.Type == "input_json_delta" {
-				tools = append(tools, model.Tool{
-					Function: model.Function{
-						Arguments: claudeResponse.Delta.PartialJson,
-					},
-				})
-			}
 		}
 	case "message_delta":
 		if claudeResponse.Usage != nil {
@ -192,10 +119,6 @@ func StreamResponseClaude2OpenAI(claudeResponse *StreamResponse) (*openai.ChatCo
 	}
 	var choice openai.ChatCompletionsStreamResponseChoice
 	choice.Delta.Content = responseText
-	if len(tools) > 0 {
-		choice.Delta.Content = nil // compatible with other OpenAI derivative applications, like LobeOpenAICompatibleFactory ...
-		choice.Delta.ToolCalls = tools
-	}
 	choice.Delta.Role = "assistant"
 	finishReason := stopReasonClaude2OpenAI(&stopReason)
 	if finishReason != "null" {
@ -212,27 +135,12 @@ func ResponseClaude2OpenAI(claudeResponse *Response) *openai.TextResponse {
 	if len(claudeResponse.Content) > 0 {
 		responseText = claudeResponse.Content[0].Text
 	}
-	tools := make([]model.Tool, 0)
-	for _, v := range claudeResponse.Content {
-		if v.Type == "tool_use" {
-			args, _ := json.Marshal(v.Input)
-			tools = append(tools, model.Tool{
-				Id:   v.Id,
-				Type: "function", // compatible with other OpenAI derivative applications
-				Function: model.Function{
-					Name:      v.Name,
-					Arguments: string(args),
-				},
-			})
-		}
-	}
 	choice := openai.TextResponseChoice{
 		Index: 0,
 		Message: model.Message{
-			Role:      "assistant",
-			Content:   responseText,
-			Name:      nil,
-			ToolCalls: tools,
+			Role:    "assistant",
+			Content: responseText,
+			Name:    nil,
 		},
 		FinishReason: stopReasonClaude2OpenAI(claudeResponse.StopReason),
 	}
@ -261,77 +169,64 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 6 {
+				continue
+			}
+			if !strings.HasPrefix(data, "data:") {
+				continue
+			}
+			data = strings.TrimPrefix(data, "data:")
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
 	var usage model.Usage
 	var modelName string
 	var id string
-	var lastToolCallChoice openai.ChatCompletionsStreamResponseChoice
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 6 || !strings.HasPrefix(data, "data:") {
-			continue
-		}
-		data = strings.TrimPrefix(data, "data:")
-		data = strings.TrimSpace(data)
-
-		var claudeResponse StreamResponse
-		err := json.Unmarshal([]byte(data), &claudeResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-
-		response, meta := StreamResponseClaude2OpenAI(&claudeResponse)
-		if meta != nil {
-			usage.PromptTokens += meta.Usage.InputTokens
-			usage.CompletionTokens += meta.Usage.OutputTokens
-			if len(meta.Id) > 0 { // only message_start has an id, otherwise it's a finish_reason event.
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			// some implementations may add \r at the end of data
+			data = strings.TrimSpace(data)
+			var claudeResponse StreamResponse
+			err := json.Unmarshal([]byte(data), &claudeResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response, meta := StreamResponseClaude2OpenAI(&claudeResponse)
+			if meta != nil {
+				usage.PromptTokens += meta.Usage.InputTokens
+				usage.CompletionTokens += meta.Usage.OutputTokens
 				modelName = meta.Model
 				id = fmt.Sprintf("chatcmpl-%s", meta.Id)
-				continue
-			} else { // finish_reason case
-				if len(lastToolCallChoice.Delta.ToolCalls) > 0 {
-					lastArgs := &lastToolCallChoice.Delta.ToolCalls[len(lastToolCallChoice.Delta.ToolCalls)-1].Function
-					if len(lastArgs.Arguments.(string)) == 0 { // compatible with OpenAI sending an empty object `{}` when no arguments.
-						lastArgs.Arguments = "{}"
-						response.Choices[len(response.Choices)-1].Delta.Content = nil
-						response.Choices[len(response.Choices)-1].Delta.ToolCalls = lastToolCallChoice.Delta.ToolCalls
-					}
-				}
+				return true
 			}
-		}
-		if response == nil {
-			continue
-		}
-
-		response.Id = id
-		response.Model = modelName
-		response.Created = createdTime
-
-		for _, choice := range response.Choices {
-			if len(choice.Delta.ToolCalls) > 0 {
-				lastToolCallChoice = choice
+			if response == nil {
+				return true
 			}
+			response.Id = id
+			response.Model = modelName
+			response.Created = createdTime
+			jsonStr, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
-	err := resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
-	}
+	})
+	_ = resp.Body.Close()
 	return nil, &usage
 }

--- a/relay/adaptor/anthropic/model.go
+++ b/relay/adaptor/anthropic/model.go
@ -16,12 +16,6 @@ type Content struct {
 	Type   string       `json:"type"`
 	Text   string       `json:"text,omitempty"`
 	Source *ImageSource `json:"source,omitempty"`
-	// tool_calls
-	Id        string `json:"id,omitempty"`
-	Name      string `json:"name,omitempty"`
-	Input     any    `json:"input,omitempty"`
-	Content   string `json:"content,omitempty"`
-	ToolUseId string `json:"tool_use_id,omitempty"`
 }

 type Message struct {
@ -29,18 +23,6 @@ type Message struct {
 	Content []Content `json:"content"`
 }

-type Tool struct {
-	Name        string      `json:"name"`
-	Description string      `json:"description,omitempty"`
-	InputSchema InputSchema `json:"input_schema"`
-}
-
-type InputSchema struct {
-	Type       string `json:"type"`
-	Properties any    `json:"properties,omitempty"`
-	Required   any    `json:"required,omitempty"`
-}
-
 type Request struct {
 	Model         string    `json:"model"`
 	Messages      []Message `json:"messages"`
@ -51,8 +33,6 @@ type Request struct {
 	Temperature   float64   `json:"temperature,omitempty"`
 	TopP          float64   `json:"top_p,omitempty"`
 	TopK          int       `json:"top_k,omitempty"`
-	Tools         []Tool    `json:"tools,omitempty"`
-	ToolChoice    any       `json:"tool_choice,omitempty"`
 	//Metadata    `json:"metadata,omitempty"`
 }

@ -81,7 +61,6 @@ type Response struct {
 type Delta struct {
 	Type         string  `json:"type"`
 	Text         string  `json:"text"`
-	PartialJson  string  `json:"partial_json,omitempty"`
 	StopReason   *string `json:"stop_reason"`
 	StopSequence *string `json:"stop_sequence"`
 }
--- a/relay/adaptor/aws/adapter.go
+++ b/relay/adaptor/aws/adapter.go
@ -1,16 +1,17 @@
 package aws

 import (
-	"errors"
-	"io"
-	"net/http"
-
 	"github.com/aws/aws-sdk-go-v2/aws"
 	"github.com/aws/aws-sdk-go-v2/credentials"
 	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
+	"github.com/songquanpeng/one-api/common/ctxkey"
+	"io"
+	"net/http"
+
 	"github.com/gin-gonic/gin"
+	"github.com/pkg/errors"
 	"github.com/songquanpeng/one-api/relay/adaptor"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
+	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
 )
@ -18,52 +19,18 @@ import (
 var _ adaptor.Adaptor = new(Adaptor)

 type Adaptor struct {
-	awsAdapter utils.AwsAdapter
-
-	Meta      *meta.Meta
-	AwsClient *bedrockruntime.Client
+	meta      *meta.Meta
+	awsClient *bedrockruntime.Client
 }

 func (a *Adaptor) Init(meta *meta.Meta) {
-	a.Meta = meta
-	a.AwsClient = bedrockruntime.New(bedrockruntime.Options{
+	a.meta = meta
+	a.awsClient = bedrockruntime.New(bedrockruntime.Options{
 		Region:      meta.Config.Region,
 		Credentials: aws.NewCredentialsCache(credentials.NewStaticCredentialsProvider(meta.Config.AK, meta.Config.SK, "")),
 	})
 }

-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	adaptor := GetAdaptor(request.Model)
-	if adaptor == nil {
-		return nil, errors.New("adaptor not found")
-	}
-
-	a.awsAdapter = adaptor
-	return adaptor.ConvertRequest(c, relayMode, request)
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	if a.awsAdapter == nil {
-		return nil, utils.WrapErr(errors.New("awsAdapter is nil"))
-	}
-	return a.awsAdapter.DoResponse(c, a.AwsClient, meta)
-}
-
-func (a *Adaptor) GetModelList() (models []string) {
-	for model := range adaptors {
-		models = append(models, model)
-	}
-	return
-}
-
-func (a *Adaptor) GetChannelName() string {
-	return "aws"
-}
-
 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
 	return "", nil
 }
@ -72,6 +39,17 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *me
 	return nil
 }

+func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+
+	claudeReq := anthropic.ConvertRequest(*request)
+	c.Set(ctxkey.RequestModel, request.Model)
+	c.Set(ctxkey.ConvertedRequest, claudeReq)
+	return claudeReq, nil
+}
+
 func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
 	if request == nil {
 		return nil, errors.New("request is nil")
@ -82,3 +60,23 @@ func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error)
 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
 	return nil, nil
 }
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
+	if meta.IsStream {
+		err, usage = StreamHandler(c, a.awsClient)
+	} else {
+		err, usage = Handler(c, a.awsClient, meta.ActualModelName)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() (models []string) {
+	for n := range awsModelIDMap {
+		models = append(models, n)
+	}
+	return
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "aws"
+}
--- a/relay/adaptor/aws/claude/adapter.go
+++ b/relay/adaptor/aws/claude/adapter.go
@ -1,37 +0,0 @@
-package aws
-
-import (
-	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-var _ utils.AwsAdapter = new(Adaptor)
-
-type Adaptor struct {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	claudeReq := anthropic.ConvertRequest(*request)
-	c.Set(ctxkey.RequestModel, request.Model)
-	c.Set(ctxkey.ConvertedRequest, claudeReq)
-	return claudeReq, nil
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, awsCli *bedrockruntime.Client, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	if meta.IsStream {
-		err, usage = StreamHandler(c, awsCli)
-	} else {
-		err, usage = Handler(c, awsCli, meta.ActualModelName)
-	}
-	return
-}
--- a/relay/adaptor/aws/llama3/adapter.go
+++ b/relay/adaptor/aws/llama3/adapter.go
@ -1,37 +0,0 @@
-package aws
-
-import (
-	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
-	"github.com/songquanpeng/one-api/common/ctxkey"
-
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-var _ utils.AwsAdapter = new(Adaptor)
-
-type Adaptor struct {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	llamaReq := ConvertRequest(*request)
-	c.Set(ctxkey.RequestModel, request.Model)
-	c.Set(ctxkey.ConvertedRequest, llamaReq)
-	return llamaReq, nil
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, awsCli *bedrockruntime.Client, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	if meta.IsStream {
-		err, usage = StreamHandler(c, awsCli)
-	} else {
-		err, usage = Handler(c, awsCli, meta.ActualModelName)
-	}
-	return
-}
--- a/relay/adaptor/aws/llama3/main.go
+++ b/relay/adaptor/aws/llama3/main.go
@ -1,231 +0,0 @@
-// Package aws provides the AWS adaptor for the relay service.
-package aws
-
-import (
-	"bytes"
-	"encoding/json"
-	"fmt"
-	"io"
-	"net/http"
-	"text/template"
-
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/common/random"
-
-	"github.com/aws/aws-sdk-go-v2/aws"
-	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
-	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime/types"
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/common"
-	"github.com/songquanpeng/one-api/common/helper"
-	"github.com/songquanpeng/one-api/common/logger"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
-	"github.com/songquanpeng/one-api/relay/adaptor/openai"
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-)
-
-// Only support llama-3-8b and llama-3-70b instruction models
-// https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html
-var AwsModelIDMap = map[string]string{
-	"llama3-8b-8192":  "meta.llama3-8b-instruct-v1:0",
-	"llama3-70b-8192": "meta.llama3-70b-instruct-v1:0",
-}
-
-func awsModelID(requestModel string) (string, error) {
-	if awsModelID, ok := AwsModelIDMap[requestModel]; ok {
-		return awsModelID, nil
-	}
-
-	return "", errors.Errorf("model %s not found", requestModel)
-}
-
-// promptTemplate with range
-const promptTemplate = `<|begin_of_text|>{{range .Messages}}<|start_header_id|>{{.Role}}<|end_header_id|>{{.StringContent}}<|eot_id|>{{end}}<|start_header_id|>assistant<|end_header_id|>
-`
-
-var promptTpl = template.Must(template.New("llama3-chat").Parse(promptTemplate))
-
-func RenderPrompt(messages []relaymodel.Message) string {
-	var buf bytes.Buffer
-	err := promptTpl.Execute(&buf, struct{ Messages []relaymodel.Message }{messages})
-	if err != nil {
-		logger.SysError("error rendering prompt messages: " + err.Error())
-	}
-	return buf.String()
-}
-
-func ConvertRequest(textRequest relaymodel.GeneralOpenAIRequest) *Request {
-	llamaRequest := Request{
-		MaxGenLen:   textRequest.MaxTokens,
-		Temperature: textRequest.Temperature,
-		TopP:        textRequest.TopP,
-	}
-	if llamaRequest.MaxGenLen == 0 {
-		llamaRequest.MaxGenLen = 2048
-	}
-	prompt := RenderPrompt(textRequest.Messages)
-	llamaRequest.Prompt = prompt
-	return &llamaRequest
-}
-
-func Handler(c *gin.Context, awsCli *bedrockruntime.Client, modelName string) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
-	awsModelId, err := awsModelID(c.GetString(ctxkey.RequestModel))
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
-	}
-
-	awsReq := &bedrockruntime.InvokeModelInput{
-		ModelId:     aws.String(awsModelId),
-		Accept:      aws.String("application/json"),
-		ContentType: aws.String("application/json"),
-	}
-
-	llamaReq, ok := c.Get(ctxkey.ConvertedRequest)
-	if !ok {
-		return utils.WrapErr(errors.New("request not found")), nil
-	}
-
-	awsReq.Body, err = json.Marshal(llamaReq)
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
-	}
-
-	awsResp, err := awsCli.InvokeModel(c.Request.Context(), awsReq)
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "InvokeModel")), nil
-	}
-
-	var llamaResponse Response
-	err = json.Unmarshal(awsResp.Body, &llamaResponse)
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "unmarshal response")), nil
-	}
-
-	openaiResp := ResponseLlama2OpenAI(&llamaResponse)
-	openaiResp.Model = modelName
-	usage := relaymodel.Usage{
-		PromptTokens:     llamaResponse.PromptTokenCount,
-		CompletionTokens: llamaResponse.GenerationTokenCount,
-		TotalTokens:      llamaResponse.PromptTokenCount + llamaResponse.GenerationTokenCount,
-	}
-	openaiResp.Usage = usage
-
-	c.JSON(http.StatusOK, openaiResp)
-	return nil, &usage
-}
-
-func ResponseLlama2OpenAI(llamaResponse *Response) *openai.TextResponse {
-	var responseText string
-	if len(llamaResponse.Generation) > 0 {
-		responseText = llamaResponse.Generation
-	}
-	choice := openai.TextResponseChoice{
-		Index: 0,
-		Message: relaymodel.Message{
-			Role:    "assistant",
-			Content: responseText,
-			Name:    nil,
-		},
-		FinishReason: llamaResponse.StopReason,
-	}
-	fullTextResponse := openai.TextResponse{
-		Id:      fmt.Sprintf("chatcmpl-%s", random.GetUUID()),
-		Object:  "chat.completion",
-		Created: helper.GetTimestamp(),
-		Choices: []openai.TextResponseChoice{choice},
-	}
-	return &fullTextResponse
-}
-
-func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
-	createdTime := helper.GetTimestamp()
-	awsModelId, err := awsModelID(c.GetString(ctxkey.RequestModel))
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
-	}
-
-	awsReq := &bedrockruntime.InvokeModelWithResponseStreamInput{
-		ModelId:     aws.String(awsModelId),
-		Accept:      aws.String("application/json"),
-		ContentType: aws.String("application/json"),
-	}
-
-	llamaReq, ok := c.Get(ctxkey.ConvertedRequest)
-	if !ok {
-		return utils.WrapErr(errors.New("request not found")), nil
-	}
-
-	awsReq.Body, err = json.Marshal(llamaReq)
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
-	}
-
-	awsResp, err := awsCli.InvokeModelWithResponseStream(c.Request.Context(), awsReq)
-	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "InvokeModelWithResponseStream")), nil
-	}
-	stream := awsResp.GetStream()
-	defer stream.Close()
-
-	c.Writer.Header().Set("Content-Type", "text/event-stream")
-	var usage relaymodel.Usage
-	c.Stream(func(w io.Writer) bool {
-		event, ok := <-stream.Events()
-		if !ok {
-			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
-			return false
-		}
-
-		switch v := event.(type) {
-		case *types.ResponseStreamMemberChunk:
-			var llamaResp StreamResponse
-			err := json.NewDecoder(bytes.NewReader(v.Value.Bytes)).Decode(&llamaResp)
-			if err != nil {
-				logger.SysError("error unmarshalling stream response: " + err.Error())
-				return false
-			}
-
-			if llamaResp.PromptTokenCount > 0 {
-				usage.PromptTokens = llamaResp.PromptTokenCount
-			}
-			if llamaResp.StopReason == "stop" {
-				usage.CompletionTokens = llamaResp.GenerationTokenCount
-				usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
-			}
-			response := StreamResponseLlama2OpenAI(&llamaResp)
-			response.Id = fmt.Sprintf("chatcmpl-%s", random.GetUUID())
-			response.Model = c.GetString(ctxkey.OriginalModel)
-			response.Created = createdTime
-			jsonStr, err := json.Marshal(response)
-			if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
-			}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
-			return true
-		case *types.UnknownUnionMember:
-			fmt.Println("unknown tag:", v.Tag)
-			return false
-		default:
-			fmt.Println("union is nil or unknown type")
-			return false
-		}
-	})
-
-	return nil, &usage
-}
-
-func StreamResponseLlama2OpenAI(llamaResponse *StreamResponse) *openai.ChatCompletionsStreamResponse {
-	var choice openai.ChatCompletionsStreamResponseChoice
-	choice.Delta.Content = llamaResponse.Generation
-	choice.Delta.Role = "assistant"
-	finishReason := llamaResponse.StopReason
-	if finishReason != "null" {
-		choice.FinishReason = &finishReason
-	}
-	var openaiResponse openai.ChatCompletionsStreamResponse
-	openaiResponse.Object = "chat.completion.chunk"
-	openaiResponse.Choices = []openai.ChatCompletionsStreamResponseChoice{choice}
-	return &openaiResponse
-}
--- a/relay/adaptor/aws/llama3/main_test.go
+++ b/relay/adaptor/aws/llama3/main_test.go
@ -1,45 +0,0 @@
-package aws_test
-
-import (
-	"testing"
-
-	aws "github.com/songquanpeng/one-api/relay/adaptor/aws/llama3"
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-	"github.com/stretchr/testify/assert"
-)
-
-func TestRenderPrompt(t *testing.T) {
-	messages := []relaymodel.Message{
-		{
-			Role:    "user",
-			Content: "What's your name?",
-		},
-	}
-	prompt := aws.RenderPrompt(messages)
-	expected := `<|begin_of_text|><|start_header_id|>user<|end_header_id|>What's your name?<|eot_id|><|start_header_id|>assistant<|end_header_id|>
-`
-	assert.Equal(t, expected, prompt)
-
-	messages = []relaymodel.Message{
-		{
-			Role:    "system",
-			Content: "Your name is Kat. You are a detective.",
-		},
-		{
-			Role:    "user",
-			Content: "What's your name?",
-		},
-		{
-			Role:    "assistant",
-			Content: "Kat",
-		},
-		{
-			Role:    "user",
-			Content: "What's your job?",
-		},
-	}
-	prompt = aws.RenderPrompt(messages)
-	expected = `<|begin_of_text|><|start_header_id|>system<|end_header_id|>Your name is Kat. You are a detective.<|eot_id|><|start_header_id|>user<|end_header_id|>What's your name?<|eot_id|><|start_header_id|>assistant<|end_header_id|>Kat<|eot_id|><|start_header_id|>user<|end_header_id|>What's your job?<|eot_id|><|start_header_id|>assistant<|end_header_id|>
-`
-	assert.Equal(t, expected, prompt)
-}
--- a/relay/adaptor/aws/llama3/model.go
+++ b/relay/adaptor/aws/llama3/model.go
@ -1,29 +0,0 @@
-package aws
-
-// Request is the request to AWS Llama3
-//
-// https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-meta.html
-type Request struct {
-	Prompt      string  `json:"prompt"`
-	MaxGenLen   int     `json:"max_gen_len,omitempty"`
-	Temperature float64 `json:"temperature,omitempty"`
-	TopP        float64 `json:"top_p,omitempty"`
-}
-
-// Response is the response from AWS Llama3
-//
-// https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-meta.html
-type Response struct {
-	Generation           string `json:"generation"`
-	PromptTokenCount     int    `json:"prompt_token_count"`
-	GenerationTokenCount int    `json:"generation_token_count"`
-	StopReason           string `json:"stop_reason"`
-}
-
-// {'generation': 'Hi', 'prompt_token_count': 15, 'generation_token_count': 1, 'stop_reason': None}
-type StreamResponse struct {
-	Generation           string `json:"generation"`
-	PromptTokenCount     int    `json:"prompt_token_count"`
-	GenerationTokenCount int    `json:"generation_token_count"`
-	StopReason           string `json:"stop_reason"`
-}
--- a/relay/adaptor/aws/claude/main.go
+++ b/relay/adaptor/aws/claude/main.go
@ -5,6 +5,7 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"io"
 	"net/http"

@ -15,28 +16,33 @@ import (
 	"github.com/jinzhu/copier"
 	"github.com/pkg/errors"
 	"github.com/songquanpeng/one-api/common"
-	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
-	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 )

+func wrapErr(err error) *relaymodel.ErrorWithStatusCode {
+	return &relaymodel.ErrorWithStatusCode{
+		StatusCode: http.StatusInternalServerError,
+		Error: relaymodel.Error{
+			Message: fmt.Sprintf("%s", err.Error()),
+		},
+	}
+}
+
 // https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html
-var AwsModelIDMap = map[string]string{
-	"claude-instant-1.2":         "anthropic.claude-instant-v1",
-	"claude-2.0":                 "anthropic.claude-v2",
-	"claude-2.1":                 "anthropic.claude-v2:1",
-	"claude-3-sonnet-20240229":   "anthropic.claude-3-sonnet-20240229-v1:0",
-	"claude-3-5-sonnet-20240620": "anthropic.claude-3-5-sonnet-20240620-v1:0",
-	"claude-3-opus-20240229":     "anthropic.claude-3-opus-20240229-v1:0",
-	"claude-3-haiku-20240307":    "anthropic.claude-3-haiku-20240307-v1:0",
+var awsModelIDMap = map[string]string{
+	"claude-instant-1.2":       "anthropic.claude-instant-v1",
+	"claude-2.0":               "anthropic.claude-v2",
+	"claude-2.1":               "anthropic.claude-v2:1",
+	"claude-3-sonnet-20240229": "anthropic.claude-3-sonnet-20240229-v1:0",
+	"claude-3-opus-20240229":   "anthropic.claude-3-opus-20240229-v1:0",
+	"claude-3-haiku-20240307":  "anthropic.claude-3-haiku-20240307-v1:0",
 }

 func awsModelID(requestModel string) (string, error) {
-	if awsModelID, ok := AwsModelIDMap[requestModel]; ok {
+	if awsModelID, ok := awsModelIDMap[requestModel]; ok {
 		return awsModelID, nil
 	}

@ -46,7 +52,7 @@ func awsModelID(requestModel string) (string, error) {
 func Handler(c *gin.Context, awsCli *bedrockruntime.Client, modelName string) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
 	awsModelId, err := awsModelID(c.GetString(ctxkey.RequestModel))
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
+		return wrapErr(errors.Wrap(err, "awsModelID")), nil
 	}

 	awsReq := &bedrockruntime.InvokeModelInput{
@ -57,30 +63,30 @@ func Handler(c *gin.Context, awsCli *bedrockruntime.Client, modelName string) (*

 	claudeReq_, ok := c.Get(ctxkey.ConvertedRequest)
 	if !ok {
-		return utils.WrapErr(errors.New("request not found")), nil
+		return wrapErr(errors.New("request not found")), nil
 	}
 	claudeReq := claudeReq_.(*anthropic.Request)
 	awsClaudeReq := &Request{
 		AnthropicVersion: "bedrock-2023-05-31",
 	}
 	if err = copier.Copy(awsClaudeReq, claudeReq); err != nil {
-		return utils.WrapErr(errors.Wrap(err, "copy request")), nil
+		return wrapErr(errors.Wrap(err, "copy request")), nil
 	}

 	awsReq.Body, err = json.Marshal(awsClaudeReq)
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
+		return wrapErr(errors.Wrap(err, "marshal request")), nil
 	}

 	awsResp, err := awsCli.InvokeModel(c.Request.Context(), awsReq)
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "InvokeModel")), nil
+		return wrapErr(errors.Wrap(err, "InvokeModel")), nil
 	}

 	claudeResponse := new(anthropic.Response)
 	err = json.Unmarshal(awsResp.Body, claudeResponse)
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "unmarshal response")), nil
+		return wrapErr(errors.Wrap(err, "unmarshal response")), nil
 	}

 	openaiResp := anthropic.ResponseClaude2OpenAI(claudeResponse)
@ -100,7 +106,7 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E
 	createdTime := helper.GetTimestamp()
 	awsModelId, err := awsModelID(c.GetString(ctxkey.RequestModel))
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
+		return wrapErr(errors.Wrap(err, "awsModelID")), nil
 	}

 	awsReq := &bedrockruntime.InvokeModelWithResponseStreamInput{
@ -111,7 +117,7 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E

 	claudeReq_, ok := c.Get(ctxkey.ConvertedRequest)
 	if !ok {
-		return utils.WrapErr(errors.New("request not found")), nil
+		return wrapErr(errors.New("request not found")), nil
 	}
 	claudeReq := claudeReq_.(*anthropic.Request)

@ -119,16 +125,16 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E
 		AnthropicVersion: "bedrock-2023-05-31",
 	}
 	if err = copier.Copy(awsClaudeReq, claudeReq); err != nil {
-		return utils.WrapErr(errors.Wrap(err, "copy request")), nil
+		return wrapErr(errors.Wrap(err, "copy request")), nil
 	}
 	awsReq.Body, err = json.Marshal(awsClaudeReq)
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
+		return wrapErr(errors.Wrap(err, "marshal request")), nil
 	}

 	awsResp, err := awsCli.InvokeModelWithResponseStream(c.Request.Context(), awsReq)
 	if err != nil {
-		return utils.WrapErr(errors.Wrap(err, "InvokeModelWithResponseStream")), nil
+		return wrapErr(errors.Wrap(err, "InvokeModelWithResponseStream")), nil
 	}
 	stream := awsResp.GetStream()
 	defer stream.Close()
@ -136,8 +142,6 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E
 	c.Writer.Header().Set("Content-Type", "text/event-stream")
 	var usage relaymodel.Usage
 	var id string
-	var lastToolCallChoice openai.ChatCompletionsStreamResponseChoice
-
 	c.Stream(func(w io.Writer) bool {
 		event, ok := <-stream.Events()
 		if !ok {
@ -158,19 +162,8 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E
 			if meta != nil {
 				usage.PromptTokens += meta.Usage.InputTokens
 				usage.CompletionTokens += meta.Usage.OutputTokens
-				if len(meta.Id) > 0 { // only message_start has an id, otherwise it's a finish_reason event.
-					id = fmt.Sprintf("chatcmpl-%s", meta.Id)
-					return true
-				} else { // finish_reason case
-					if len(lastToolCallChoice.Delta.ToolCalls) > 0 {
-						lastArgs := &lastToolCallChoice.Delta.ToolCalls[len(lastToolCallChoice.Delta.ToolCalls)-1].Function
-						if len(lastArgs.Arguments.(string)) == 0 { // compatible with OpenAI sending an empty object `{}` when no arguments.
-							lastArgs.Arguments = "{}"
-							response.Choices[len(response.Choices)-1].Delta.Content = nil
-							response.Choices[len(response.Choices)-1].Delta.ToolCalls = lastToolCallChoice.Delta.ToolCalls
-						}
-					}
-				}
+				id = fmt.Sprintf("chatcmpl-%s", meta.Id)
+				return true
 			}
 			if response == nil {
 				return true
@ -178,12 +171,6 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E
 			response.Id = id
 			response.Model = c.GetString(ctxkey.OriginalModel)
 			response.Created = createdTime
-
-			for _, choice := range response.Choices {
-				if len(choice.Delta.ToolCalls) > 0 {
-					lastToolCallChoice = choice
-				}
-			}
 			jsonStr, err := json.Marshal(response)
 			if err != nil {
 				logger.SysError("error marshalling stream response: " + err.Error())
--- a/relay/adaptor/aws/claude/model.go
+++ b/relay/adaptor/aws/claude/model.go
@ -9,12 +9,9 @@ type Request struct {
 	// AnthropicVersion should be "bedrock-2023-05-31"
 	AnthropicVersion string              `json:"anthropic_version"`
 	Messages         []anthropic.Message `json:"messages"`
-	System           string              `json:"system,omitempty"`
 	MaxTokens        int                 `json:"max_tokens,omitempty"`
 	Temperature      float64             `json:"temperature,omitempty"`
 	TopP             float64             `json:"top_p,omitempty"`
 	TopK             int                 `json:"top_k,omitempty"`
 	StopSequences    []string            `json:"stop_sequences,omitempty"`
-	Tools            []anthropic.Tool    `json:"tools,omitempty"`
-	ToolChoice       any                 `json:"tool_choice,omitempty"`
 }
--- a/relay/adaptor/aws/registry.go
+++ b/relay/adaptor/aws/registry.go
@ -1,39 +0,0 @@
-package aws
-
-import (
-	claude "github.com/songquanpeng/one-api/relay/adaptor/aws/claude"
-	llama3 "github.com/songquanpeng/one-api/relay/adaptor/aws/llama3"
-	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
-)
-
-type AwsModelType int
-
-const (
-	AwsClaude AwsModelType = iota + 1
-	AwsLlama3
-)
-
-var (
-	adaptors = map[string]AwsModelType{}
-)
-
-func init() {
-	for model := range claude.AwsModelIDMap {
-		adaptors[model] = AwsClaude
-	}
-	for model := range llama3.AwsModelIDMap {
-		adaptors[model] = AwsLlama3
-	}
-}
-
-func GetAdaptor(model string) utils.AwsAdapter {
-	adaptorType := adaptors[model]
-	switch adaptorType {
-	case AwsClaude:
-		return &claude.Adaptor{}
-	case AwsLlama3:
-		return &llama3.Adaptor{}
-	default:
-		return nil
-	}
-}
--- a/relay/adaptor/aws/utils/adaptor.go
+++ b/relay/adaptor/aws/utils/adaptor.go
@ -1,51 +0,0 @@
-package utils
-
-import (
-	"errors"
-	"io"
-	"net/http"
-
-	"github.com/aws/aws-sdk-go-v2/aws"
-	"github.com/aws/aws-sdk-go-v2/credentials"
-	"github.com/aws/aws-sdk-go-v2/service/bedrockruntime"
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-type AwsAdapter interface {
-	ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error)
-	DoResponse(c *gin.Context, awsCli *bedrockruntime.Client, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode)
-}
-
-type Adaptor struct {
-	Meta      *meta.Meta
-	AwsClient *bedrockruntime.Client
-}
-
-func (a *Adaptor) Init(meta *meta.Meta) {
-	a.Meta = meta
-	a.AwsClient = bedrockruntime.New(bedrockruntime.Options{
-		Region:      meta.Config.Region,
-		Credentials: aws.NewCredentialsCache(credentials.NewStaticCredentialsProvider(meta.Config.AK, meta.Config.SK, "")),
-	})
-}
-
-func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	return "", nil
-}
-
-func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
-	return nil
-}
-
-func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-	return request, nil
-}
-
-func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
-	return nil, nil
-}
--- a/relay/adaptor/aws/utils/utils.go
+++ b/relay/adaptor/aws/utils/utils.go
@ -1,16 +0,0 @@
-package utils
-
-import (
-	"net/http"
-
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-)
-
-func WrapErr(err error) *relaymodel.ErrorWithStatusCode {
-	return &relaymodel.ErrorWithStatusCode{
-		StatusCode: http.StatusInternalServerError,
-		Error: relaymodel.Error{
-			Message: err.Error(),
-		},
-	}
-}
--- a/relay/adaptor/baidu/main.go
+++ b/relay/adaptor/baidu/main.go
@ -5,13 +5,6 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strings"
-	"sync"
-	"time"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/client"
@ -19,6 +12,11 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
+	"sync"
+	"time"
 )

 // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/flfmc9do2
@ -139,41 +137,59 @@ func embeddingResponseBaidu2OpenAI(response *EmbeddingResponse) *openai.Embeddin
 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	var usage model.Usage
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
-
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "\n"); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 6 { // ignore blank line or wrong format
+				continue
+			}
+			data = data[6:]
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 6 {
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var baiduResponse ChatStreamResponse
+			err := json.Unmarshal([]byte(data), &baiduResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			if baiduResponse.Usage.TotalTokens != 0 {
+				usage.TotalTokens = baiduResponse.Usage.TotalTokens
+				usage.PromptTokens = baiduResponse.Usage.PromptTokens
+				usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens
+			}
+			response := streamResponseBaidu2OpenAI(&baiduResponse)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = data[6:]
-
-		var baiduResponse ChatStreamResponse
-		err := json.Unmarshal([]byte(data), &baiduResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-		if baiduResponse.Usage.TotalTokens != 0 {
-			usage.TotalTokens = baiduResponse.Usage.TotalTokens
-			usage.PromptTokens = baiduResponse.Usage.PromptTokens
-			usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens
-		}
-		response := streamResponseBaidu2OpenAI(&baiduResponse)
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
--- a/relay/adaptor/cloudflare/adaptor.go
+++ b/relay/adaptor/cloudflare/adaptor.go
@ -5,13 +5,11 @@ import (
 	"fmt"
 	"io"
 	"net/http"
-	"strings"

 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
-	"github.com/songquanpeng/one-api/relay/relaymode"
 )

 type Adaptor struct {
@ -29,33 +27,8 @@ func (a *Adaptor) Init(meta *meta.Meta) {
 	a.meta = meta
 }

-// WorkerAI cannot be used across accounts with AIGateWay
-// https://developers.cloudflare.com/ai-gateway/providers/workersai/#openai-compatible-endpoints
-// https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id}/workers-ai
-func (a *Adaptor) isAIGateWay(baseURL string) bool {
-	return strings.HasPrefix(baseURL, "https://gateway.ai.cloudflare.com") && strings.HasSuffix(baseURL, "/workers-ai")
-}
-
 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	isAIGateWay := a.isAIGateWay(meta.BaseURL)
-	var urlPrefix string
-	if isAIGateWay {
-		urlPrefix = meta.BaseURL
-	} else {
-		urlPrefix = fmt.Sprintf("%s/client/v4/accounts/%s/ai", meta.BaseURL, meta.Config.UserID)
-	}
-
-	switch meta.Mode {
-	case relaymode.ChatCompletions:
-		return fmt.Sprintf("%s/v1/chat/completions", urlPrefix), nil
-	case relaymode.Embeddings:
-		return fmt.Sprintf("%s/v1/embeddings", urlPrefix), nil
-	default:
-		if isAIGateWay {
-			return fmt.Sprintf("%s/%s", urlPrefix, meta.ActualModelName), nil
-		}
-		return fmt.Sprintf("%s/run/%s", urlPrefix, meta.ActualModelName), nil
-	}
+	return fmt.Sprintf("%s/client/v4/accounts/%s/ai/run/%s", meta.BaseURL, meta.Config.UserID, meta.ActualModelName), nil
 }

 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
@ -68,14 +41,7 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
-	switch relayMode {
-	case relaymode.Completions:
-		return ConvertCompletionsRequest(*request), nil
-	case relaymode.ChatCompletions, relaymode.Embeddings:
-		return request, nil
-	default:
-		return nil, errors.New("not implemented")
-	}
+	return ConvertRequest(*request), nil
 }

 func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
--- a/relay/adaptor/cloudflare/constant.go
+++ b/relay/adaptor/cloudflare/constant.go
@ -1,7 +1,6 @@
 package cloudflare

 var ModelList = []string{
-	"@cf/meta/llama-3.1-8b-instruct",
 	"@cf/meta/llama-2-7b-chat-fp16",
 	"@cf/meta/llama-2-7b-chat-int8",
 	"@cf/mistral/mistral-7b-instruct-v0.1",
--- a/relay/adaptor/cloudflare/main.go
+++ b/relay/adaptor/cloudflare/main.go
@ -2,14 +2,12 @@ package cloudflare

 import (
 	"bufio"
+	"bytes"
 	"encoding/json"
 	"io"
 	"net/http"
 	"strings"

-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/common/render"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
@ -18,66 +16,114 @@ import (
 	"github.com/songquanpeng/one-api/relay/model"
 )

-func ConvertCompletionsRequest(textRequest model.GeneralOpenAIRequest) *Request {
-	p, _ := textRequest.Prompt.(string)
-	return &Request{
-		Prompt:      p,
-		MaxTokens:   textRequest.MaxTokens,
-		Stream:      textRequest.Stream,
-		Temperature: textRequest.Temperature,
+func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
+    var promptBuilder strings.Builder
+    for _, message := range textRequest.Messages {
+        promptBuilder.WriteString(message.StringContent())
+        promptBuilder.WriteString("\n")  // 添加换行符来分隔每个消息
+    }
+
+    return &Request{
+        MaxTokens:   textRequest.MaxTokens,
+        Prompt:      promptBuilder.String(),
+        Stream:      textRequest.Stream,
+        Temperature: textRequest.Temperature,
+    }
+}
+
+
+func ResponseCloudflare2OpenAI(cloudflareResponse *Response) *openai.TextResponse {
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    "assistant",
+			Content: cloudflareResponse.Result.Response,
+		},
+		FinishReason: "stop",
 	}
+	fullTextResponse := openai.TextResponse{
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+	}
+	return &fullTextResponse
+}
+
+func StreamResponseCloudflare2OpenAI(cloudflareResponse *StreamResponse) *openai.ChatCompletionsStreamResponse {
+	var choice openai.ChatCompletionsStreamResponseChoice
+	choice.Delta.Content = cloudflareResponse.Response
+	choice.Delta.Role = "assistant"
+	openaiResponse := openai.ChatCompletionsStreamResponse{
+		Object:  "chat.completion.chunk",
+		Choices: []openai.ChatCompletionsStreamResponseChoice{choice},
+		Created: helper.GetTimestamp(),
+	}
+	return &openaiResponse
 }

 func StreamHandler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *model.Usage) {
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := bytes.IndexByte(data, '\n'); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})

+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < len("data: ") {
+				continue
+			}
+			data = strings.TrimPrefix(data, "data: ")
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
 	id := helper.GetResponseID(c)
-	responseModel := c.GetString(ctxkey.OriginalModel)
+	responseModel := c.GetString("original_model")
 	var responseText string
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < len("data: ") {
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			// some implementations may add \r at the end of data
+			data = strings.TrimSuffix(data, "\r")
+			var cloudflareResponse StreamResponse
+			err := json.Unmarshal([]byte(data), &cloudflareResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response := StreamResponseCloudflare2OpenAI(&cloudflareResponse)
+			if response == nil {
+				return true
+			}
+			responseText += cloudflareResponse.Response
+			response.Id = id
+			response.Model = responseModel
+			jsonStr, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = strings.TrimPrefix(data, "data: ")
-		data = strings.TrimSuffix(data, "\r")
-
-		if data == "[DONE]" {
-			break
-		}
-
-		var response openai.ChatCompletionsStreamResponse
-		err := json.Unmarshal([]byte(data), &response)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-		for _, v := range response.Choices {
-			v.Delta.Role = "assistant"
-			responseText += v.Delta.StringContent()
-		}
-		response.Id = id
-		response.Model = modelName
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
-	err := resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
-	}
-
+	})
+	_ = resp.Body.Close()
 	usage := openai.ResponseText2Usage(responseText, responseModel, promptTokens)
 	return nil, usage
 }
@ -91,25 +137,22 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
-	var response openai.TextResponse
-	err = json.Unmarshal(responseBody, &response)
+	var cloudflareResponse Response
+	err = json.Unmarshal(responseBody, &cloudflareResponse)
 	if err != nil {
 		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
 	}
-	response.Model = modelName
-	var responseText string
-	for _, v := range response.Choices {
-		responseText += v.Message.Content.(string)
-	}
-	usage := openai.ResponseText2Usage(responseText, modelName, promptTokens)
-	response.Usage = *usage
-	response.Id = helper.GetResponseID(c)
-	jsonResponse, err := json.Marshal(response)
+	fullTextResponse := ResponseCloudflare2OpenAI(&cloudflareResponse)
+	fullTextResponse.Model = modelName
+	usage := openai.ResponseText2Usage(cloudflareResponse.Result.Response, modelName, promptTokens)
+	fullTextResponse.Usage = *usage
+	fullTextResponse.Id = helper.GetResponseID(c)
+	jsonResponse, err := json.Marshal(fullTextResponse)
 	if err != nil {
 		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
 	}
 	c.Writer.Header().Set("Content-Type", "application/json")
 	c.Writer.WriteHeader(resp.StatusCode)
-	_, _ = c.Writer.Write(jsonResponse)
+	_, err = c.Writer.Write(jsonResponse)
 	return nil, usage
 }
--- a/relay/adaptor/cloudflare/model.go
+++ b/relay/adaptor/cloudflare/model.go
@ -1,13 +1,25 @@
 package cloudflare

-import "github.com/songquanpeng/one-api/relay/model"
-
 type Request struct {
-	Messages    []model.Message `json:"messages,omitempty"`
-	Lora        string          `json:"lora,omitempty"`
-	MaxTokens   int             `json:"max_tokens,omitempty"`
-	Prompt      string          `json:"prompt,omitempty"`
-	Raw         bool            `json:"raw,omitempty"`
-	Stream      bool            `json:"stream,omitempty"`
-	Temperature float64         `json:"temperature,omitempty"`
+	Lora        string  `json:"lora,omitempty"`
+	MaxTokens   int     `json:"max_tokens,omitempty"`
+	Prompt      string  `json:"prompt,omitempty"`
+	Raw         bool    `json:"raw,omitempty"`
+	Stream      bool    `json:"stream,omitempty"`
+	Temperature float64 `json:"temperature,omitempty"`
+}
+
+type Result struct {
+	Response string `json:"response"`
+}
+
+type Response struct {
+	Result   Result   `json:"result"`
+	Success  bool     `json:"success"`
+	Errors   []string `json:"errors"`
+	Messages []string `json:"messages"`
+}
+
+type StreamResponse struct {
+	Response string `json:"response"`
 }
--- a/relay/adaptor/cohere/main.go
+++ b/relay/adaptor/cohere/main.go
@ -2,9 +2,9 @@ package cohere

 import (
 	"bufio"
+	"bytes"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
 	"io"
 	"net/http"
 	"strings"
@ -134,53 +134,66 @@ func ResponseCohere2OpenAI(cohereResponse *Response) *openai.TextResponse {
 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
 	createdTime := helper.GetTimestamp()
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := bytes.IndexByte(data, '\n'); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})

+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
 	var usage model.Usage
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		data = strings.TrimSuffix(data, "\r")
-
-		var cohereResponse StreamResponse
-		err := json.Unmarshal([]byte(data), &cohereResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			// some implementations may add \r at the end of data
+			data = strings.TrimSuffix(data, "\r")
+			var cohereResponse StreamResponse
+			err := json.Unmarshal([]byte(data), &cohereResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response, meta := StreamResponseCohere2OpenAI(&cohereResponse)
+			if meta != nil {
+				usage.PromptTokens += meta.Meta.Tokens.InputTokens
+				usage.CompletionTokens += meta.Meta.Tokens.OutputTokens
+				return true
+			}
+			if response == nil {
+				return true
+			}
+			response.Id = fmt.Sprintf("chatcmpl-%d", createdTime)
+			response.Model = c.GetString("original_model")
+			response.Created = createdTime
+			jsonStr, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-
-		response, meta := StreamResponseCohere2OpenAI(&cohereResponse)
-		if meta != nil {
-			usage.PromptTokens += meta.Meta.Tokens.InputTokens
-			usage.CompletionTokens += meta.Meta.Tokens.OutputTokens
-			continue
-		}
-		if response == nil {
-			continue
-		}
-
-		response.Id = fmt.Sprintf("chatcmpl-%d", createdTime)
-		response.Model = c.GetString("original_model")
-		response.Created = createdTime
-
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
-	err := resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
-	}
-
+	})
+	_ = resp.Body.Close()
 	return nil, &usage
 }

--- a/relay/adaptor/coze/main.go
+++ b/relay/adaptor/coze/main.go
@ -4,11 +4,6 @@ import (
 	"bufio"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/conv"
@ -17,6 +12,9 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/coze/constant/messagetype"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
 )

 // https://www.coze.com/open
@ -111,54 +109,69 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 	var responseText string
 	createdTime := helper.GetTimestamp()
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
-
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "\n"); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 5 {
+				continue
+			}
+			if !strings.HasPrefix(data, "data:") {
+				continue
+			}
+			data = strings.TrimPrefix(data, "data:")
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
 	var modelName string
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 5 || !strings.HasPrefix(data, "data:") {
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			// some implementations may add \r at the end of data
+			data = strings.TrimSuffix(data, "\r")
+			var cozeResponse StreamResponse
+			err := json.Unmarshal([]byte(data), &cozeResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response, _ := StreamResponseCoze2OpenAI(&cozeResponse)
+			if response == nil {
+				return true
+			}
+			for _, choice := range response.Choices {
+				responseText += conv.AsString(choice.Delta.Content)
+			}
+			response.Model = modelName
+			response.Created = createdTime
+			jsonStr, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = strings.TrimPrefix(data, "data:")
-		data = strings.TrimSuffix(data, "\r")
-
-		var cozeResponse StreamResponse
-		err := json.Unmarshal([]byte(data), &cozeResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-
-		response, _ := StreamResponseCoze2OpenAI(&cozeResponse)
-		if response == nil {
-			continue
-		}
-
-		for _, choice := range response.Choices {
-			responseText += conv.AsString(choice.Delta.Content)
-		}
-		response.Model = modelName
-		response.Created = createdTime
-
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
-	err := resp.Body.Close()
-	if err != nil {
-		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
-	}
-
+	})
+	_ = resp.Body.Close()
 	return nil, &responseText
 }

--- a/relay/adaptor/doubao/main.go
+++ b/relay/adaptor/doubao/main.go
@ -7,12 +7,8 @@ import (
 )

 func GetRequestURL(meta *meta.Meta) (string, error) {
-	switch meta.Mode {
-	case relaymode.ChatCompletions:
+	if meta.Mode == relaymode.ChatCompletions {
 		return fmt.Sprintf("%s/api/v3/chat/completions", meta.BaseURL), nil
-	case relaymode.Embeddings:
-		return fmt.Sprintf("%s/api/v3/embeddings", meta.BaseURL), nil
-	default:
 	}
 	return "", fmt.Errorf("unsupported relay mode %d for doubao", meta.Mode)
 }
--- a/relay/adaptor/gemini/constants.go
+++ b/relay/adaptor/gemini/constants.go
@ -3,5 +3,6 @@ package gemini
 // https://ai.google.dev/models/gemini

 var ModelList = []string{
-	"gemini-pro", "gemini-1.0-pro", "gemini-1.5-flash", "gemini-1.5-pro", "text-embedding-004", "aqa",
+	"gemini-pro", "gemini-1.0-pro-001", "gemini-1.5-pro",
+	"gemini-pro-vision", "gemini-1.0-pro-vision-001", "embedding-001", "text-embedding-004",
 }
--- a/relay/adaptor/gemini/main.go
+++ b/relay/adaptor/gemini/main.go
@ -4,7 +4,6 @@ import (
 	"bufio"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
 	"io"
 	"net/http"
 	"strings"
@ -276,50 +275,64 @@ func embeddingResponseGemini2OpenAI(response *EmbeddingResponse) *openai.Embeddi
 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, string) {
 	responseText := ""
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
-
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "\n"); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			data = strings.TrimSpace(data)
+			if !strings.HasPrefix(data, "data: ") {
+				continue
+			}
+			data = strings.TrimPrefix(data, "data: ")
+			data = strings.TrimSuffix(data, "\"")
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		data = strings.TrimSpace(data)
-		if !strings.HasPrefix(data, "data: ") {
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var geminiResponse ChatResponse
+			err := json.Unmarshal([]byte(data), &geminiResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response := streamResponseGeminiChat2OpenAI(&geminiResponse)
+			if response == nil {
+				return true
+			}
+			responseText += response.Choices[0].Delta.StringContent()
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = strings.TrimPrefix(data, "data: ")
-		data = strings.TrimSuffix(data, "\"")
-
-		var geminiResponse ChatResponse
-		err := json.Unmarshal([]byte(data), &geminiResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-
-		response := streamResponseGeminiChat2OpenAI(&geminiResponse)
-		if response == nil {
-			continue
-		}
-
-		responseText += response.Choices[0].Delta.StringContent()
-
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), ""
 	}
-
 	return nil, responseText
 }

--- a/relay/adaptor/groq/constants.go
+++ b/relay/adaptor/groq/constants.go
@ -4,14 +4,9 @@ package groq

 var ModelList = []string{
 	"gemma-7b-it",
+	"llama2-7b-2048",
+	"llama2-70b-4096",
 	"mixtral-8x7b-32768",
 	"llama3-8b-8192",
 	"llama3-70b-8192",
-	"gemma2-9b-it",
-	"llama-3.1-405b-reasoning",
-	"llama-3.1-70b-versatile",
-	"llama-3.1-8b-instant",
-	"llama3-groq-70b-8192-tool-use-preview",
-	"llama3-groq-8b-8192-tool-use-preview",
-	"whisper-large-v3",
 }
--- a/relay/adaptor/novita/constants.go
+++ b/relay/adaptor/novita/constants.go
@ -1,19 +0,0 @@
-package novita
-
-// https://novita.ai/llm-api
-
-var ModelList = []string{
-	"meta-llama/llama-3-8b-instruct",
-	"meta-llama/llama-3-70b-instruct",
-	"nousresearch/hermes-2-pro-llama-3-8b",
-	"nousresearch/nous-hermes-llama2-13b",
-	"mistralai/mistral-7b-instruct",
-	"cognitivecomputations/dolphin-mixtral-8x22b",
-	"sao10k/l3-70b-euryale-v2.1",
-	"sophosympatheia/midnight-rose-70b",
-	"gryphe/mythomax-l2-13b",
-	"Nous-Hermes-2-Mixtral-8x7B-DPO",
-	"lzlv_70b",
-	"teknium/openhermes-2.5-mistral-7b",
-	"microsoft/wizardlm-2-8x22b",
-}
--- a/relay/adaptor/novita/main.go
+++ b/relay/adaptor/novita/main.go
@ -1,15 +0,0 @@
-package novita
-
-import (
-	"fmt"
-
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/relaymode"
-)
-
-func GetRequestURL(meta *meta.Meta) (string, error) {
-	if meta.Mode == relaymode.ChatCompletions {
-		return fmt.Sprintf("%s/chat/completions", meta.BaseURL), nil
-	}
-	return "", fmt.Errorf("unsupported relay mode %d for novita", meta.Mode)
-}
--- a/relay/adaptor/ollama/adaptor.go
+++ b/relay/adaptor/ollama/adaptor.go
@ -24,7 +24,7 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
 	// https://github.com/ollama/ollama/blob/main/docs/api.md
 	fullRequestURL := fmt.Sprintf("%s/api/chat", meta.BaseURL)
 	if meta.Mode == relaymode.Embeddings {
-		fullRequestURL = fmt.Sprintf("%s/api/embed", meta.BaseURL)
+		fullRequestURL = fmt.Sprintf("%s/api/embeddings", meta.BaseURL)
 	}
 	return fullRequestURL, nil
 }
--- a/relay/adaptor/ollama/main.go
+++ b/relay/adaptor/ollama/main.go
@ -5,14 +5,12 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/random"
 	"io"
 	"net/http"
 	"strings"

-	"github.com/songquanpeng/one-api/common/helper"
-	"github.com/songquanpeng/one-api/common/random"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/image"
@ -31,8 +29,6 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 			TopP:             request.TopP,
 			FrequencyPenalty: request.FrequencyPenalty,
 			PresencePenalty:  request.PresencePenalty,
-			NumPredict:  	  request.MaxTokens,
-			NumCtx:  	  request.NumCtx,
 		},
 		Stream: request.Stream,
 	}
@ -109,67 +105,61 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 			return 0, nil, nil
 		}
 		if i := strings.Index(string(data), "}\n"); i >= 0 {
-			return i + 2, data[0 : i+1], nil
+			return i + 2, data[0:i], nil
 		}
 		if atEOF {
 			return len(data), data, nil
 		}
 		return 0, nil, nil
 	})
-
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := strings.TrimPrefix(scanner.Text(), "}")
+			dataChan <- data + "}"
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if strings.HasPrefix(data, "}") {
-		    data = strings.TrimPrefix(data, "}") + "}"
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var ollamaResponse ChatResponse
+			err := json.Unmarshal([]byte(data), &ollamaResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			if ollamaResponse.EvalCount != 0 {
+				usage.PromptTokens = ollamaResponse.PromptEvalCount
+				usage.CompletionTokens = ollamaResponse.EvalCount
+				usage.TotalTokens = ollamaResponse.PromptEvalCount + ollamaResponse.EvalCount
+			}
+			response := streamResponseOllama2OpenAI(&ollamaResponse)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-
-		var ollamaResponse ChatResponse
-		err := json.Unmarshal([]byte(data), &ollamaResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-
-		if ollamaResponse.EvalCount != 0 {
-			usage.PromptTokens = ollamaResponse.PromptEvalCount
-			usage.CompletionTokens = ollamaResponse.EvalCount
-			usage.TotalTokens = ollamaResponse.PromptEvalCount + ollamaResponse.EvalCount
-		}
-
-		response := streamResponseOllama2OpenAI(&ollamaResponse)
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
-
 	return nil, &usage
 }

 func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
 	return &EmbeddingRequest{
-		Model: request.Model,
-		Input: request.ParseInput(),
-		Options: &Options{
-			Seed:             int(request.Seed),
-			Temperature:      request.Temperature,
-			TopP:             request.TopP,
-			FrequencyPenalty: request.FrequencyPenalty,
-			PresencePenalty:  request.PresencePenalty,
-		},
+		Model:  request.Model,
+		Prompt: strings.Join(request.ParseInput(), " "),
 	}
 }

@ -212,17 +202,15 @@ func embeddingResponseOllama2OpenAI(response *EmbeddingResponse) *openai.Embeddi
 	openAIEmbeddingResponse := openai.EmbeddingResponse{
 		Object: "list",
 		Data:   make([]openai.EmbeddingResponseItem, 0, 1),
-		Model:  response.Model,
+		Model:  "text-embedding-v1",
 		Usage:  model.Usage{TotalTokens: 0},
 	}

-	for i, embedding := range response.Embeddings {
-		openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{
-			Object:    `embedding`,
-			Index:     i,
-			Embedding: embedding,
-		})
-	}
+	openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{
+		Object:    `embedding`,
+		Index:     0,
+		Embedding: response.Embedding,
+	})
 	return &openAIEmbeddingResponse
 }

--- a/relay/adaptor/ollama/model.go
+++ b/relay/adaptor/ollama/model.go
@ -7,8 +7,6 @@ type Options struct {
 	TopP             float64 `json:"top_p,omitempty"`
 	FrequencyPenalty float64 `json:"frequency_penalty,omitempty"`
 	PresencePenalty  float64 `json:"presence_penalty,omitempty"`
-	NumPredict  	 int 	 `json:"num_predict,omitempty"`
-	NumCtx  	 int 	 `json:"num_ctx,omitempty"`
 }

 type Message struct {
@ -39,15 +37,11 @@ type ChatResponse struct {
 }

 type EmbeddingRequest struct {
-	Model string   `json:"model"`
-	Input []string `json:"input"`
-	// Truncate  bool     `json:"truncate,omitempty"`
-	Options *Options `json:"options,omitempty"`
-	// KeepAlive string   `json:"keep_alive,omitempty"`
+	Model  string `json:"model"`
+	Prompt string `json:"prompt"`
 }

 type EmbeddingResponse struct {
-	Error      string      `json:"error,omitempty"`
-	Model      string      `json:"model"`
-	Embeddings [][]float64 `json:"embeddings"`
+	Error     string    `json:"error,omitempty"`
+	Embedding []float64 `json:"embedding,omitempty"`
 }
--- a/relay/adaptor/openai/adaptor.go
+++ b/relay/adaptor/openai/adaptor.go
@ -3,19 +3,17 @@ package openai
 import (
 	"errors"
 	"fmt"
-	"io"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/adaptor/doubao"
 	"github.com/songquanpeng/one-api/relay/adaptor/minimax"
-	"github.com/songquanpeng/one-api/relay/adaptor/novita"
 	"github.com/songquanpeng/one-api/relay/channeltype"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
+	"io"
+	"net/http"
+	"strings"
 )

 type Adaptor struct {
@ -50,8 +48,6 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
 		return minimax.GetRequestURL(meta)
 	case channeltype.Doubao:
 		return doubao.GetRequestURL(meta)
-	case channeltype.Novita:
-		return novita.GetRequestURL(meta)
 	default:
 		return GetFullRequestURL(meta.BaseURL, meta.RequestURLPath, meta.ChannelType), nil
 	}
--- a/relay/adaptor/openai/compatible.go
+++ b/relay/adaptor/openai/compatible.go
@ -10,10 +10,8 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/minimax"
 	"github.com/songquanpeng/one-api/relay/adaptor/mistral"
 	"github.com/songquanpeng/one-api/relay/adaptor/moonshot"
-	"github.com/songquanpeng/one-api/relay/adaptor/novita"
 	"github.com/songquanpeng/one-api/relay/adaptor/stepfun"
 	"github.com/songquanpeng/one-api/relay/adaptor/togetherai"
-	"github.com/songquanpeng/one-api/relay/adaptor/siliconflow"
 	"github.com/songquanpeng/one-api/relay/channeltype"
 )

@ -30,8 +28,6 @@ var CompatibleChannels = []int{
 	channeltype.StepFun,
 	channeltype.DeepSeek,
 	channeltype.TogetherAI,
-	channeltype.Novita,
-	channeltype.SiliconFlow,
 }

 func GetCompatibleChannelMeta(channelType int) (string, []string) {
@ -60,10 +56,6 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) {
 		return "together.ai", togetherai.ModelList
 	case channeltype.Doubao:
 		return "doubao", doubao.ModelList
-	case channeltype.Novita:
-		return "novita", novita.ModelList
-	case channeltype.SiliconFlow:
-		return "siliconflow", siliconflow.ModelList
 	default:
 		return "openai", ModelList
 	}
--- a/relay/adaptor/openai/constants.go
+++ b/relay/adaptor/openai/constants.go
@ -8,9 +8,6 @@ var ModelList = []string{
 	"gpt-4-32k", "gpt-4-32k-0314", "gpt-4-32k-0613",
 	"gpt-4-turbo-preview", "gpt-4-turbo", "gpt-4-turbo-2024-04-09",
 	"gpt-4o", "gpt-4o-2024-05-13",
-	"gpt-4o-2024-08-06",
-	"chatgpt-4o-latest",
-	"gpt-4o-mini", "gpt-4o-mini-2024-07-18",
 	"gpt-4-vision-preview",
 	"text-embedding-ada-002", "text-embedding-3-small", "text-embedding-3-large",
 	"text-curie-001", "text-babbage-001", "text-ada-001", "text-davinci-002", "text-davinci-003",
--- a/relay/adaptor/openai/main.go
+++ b/relay/adaptor/openai/main.go
@ -4,18 +4,15 @@ import (
 	"bufio"
 	"bytes"
 	"encoding/json"
-	"io"
-	"net/http"
-	"strings"
-
-	"github.com/songquanpeng/one-api/common/render"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/conv"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
+	"io"
+	"net/http"
+	"strings"
 )

 const (
@ -27,72 +24,88 @@ const (
 func StreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*model.ErrorWithStatusCode, string, *model.Usage) {
 	responseText := ""
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "\n"); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
 	var usage *model.Usage
-
-	common.SetEventStreamHeaders(c)
-
-	doneRendered := false
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < dataPrefixLength { // ignore blank line or wrong format
-			continue
-		}
-		if data[:dataPrefixLength] != dataPrefix && data[:dataPrefixLength] != done {
-			continue
-		}
-		if strings.HasPrefix(data[dataPrefixLength:], done) {
-			render.StringData(c, data)
-			doneRendered = true
-			continue
-		}
-		switch relayMode {
-		case relaymode.ChatCompletions:
-			var streamResponse ChatCompletionsStreamResponse
-			err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse)
-			if err != nil {
-				logger.SysError("error unmarshalling stream response: " + err.Error())
-				render.StringData(c, data) // if error happened, pass the data to client
-				continue                   // just ignore the error
-			}
-			if len(streamResponse.Choices) == 0 && streamResponse.Usage == nil {
-				// but for empty choice and no usage, we should not pass it to client, this is for azure
-				continue // just ignore empty choice
-			}
-			render.StringData(c, data)
-			for _, choice := range streamResponse.Choices {
-				responseText += conv.AsString(choice.Delta.Content)
-			}
-			if streamResponse.Usage != nil {
-				usage = streamResponse.Usage
-			}
-		case relaymode.Completions:
-			render.StringData(c, data)
-			var streamResponse CompletionsStreamResponse
-			err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse)
-			if err != nil {
-				logger.SysError("error unmarshalling stream response: " + err.Error())
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < dataPrefixLength { // ignore blank line or wrong format
 				continue
 			}
-			for _, choice := range streamResponse.Choices {
-				responseText += choice.Text
+			if data[:dataPrefixLength] != dataPrefix && data[:dataPrefixLength] != done {
+				continue
+			}
+			if strings.HasPrefix(data[dataPrefixLength:], done) {
+				dataChan <- data
+				continue
+			}
+			switch relayMode {
+			case relaymode.ChatCompletions:
+				var streamResponse ChatCompletionsStreamResponse
+				err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse)
+				if err != nil {
+					logger.SysError("error unmarshalling stream response: " + err.Error())
+					dataChan <- data // if error happened, pass the data to client
+					continue         // just ignore the error
+				}
+				if len(streamResponse.Choices) == 0 {
+					// but for empty choice, we should not pass it to client, this is for azure
+					continue // just ignore empty choice
+				}
+				dataChan <- data
+				for _, choice := range streamResponse.Choices {
+					responseText += conv.AsString(choice.Delta.Content)
+				}
+				if streamResponse.Usage != nil {
+					usage = streamResponse.Usage
+				}
+			case relaymode.Completions:
+				dataChan <- data
+				var streamResponse CompletionsStreamResponse
+				err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse)
+				if err != nil {
+					logger.SysError("error unmarshalling stream response: " + err.Error())
+					continue
+				}
+				for _, choice := range streamResponse.Choices {
+					responseText += choice.Text
+				}
 			}
 		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	if !doneRendered {
-		render.Done(c)
-	}
-
+		stopChan <- true
+	}()
+	common.SetEventStreamHeaders(c)
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			if strings.HasPrefix(data, "data: [DONE]") {
+				data = data[:12]
+			}
+			// some implementations may add \r at the end of data
+			data = strings.TrimSuffix(data, "\r")
+			c.Render(-1, common.CustomEvent{Data: data})
+			return true
+		case <-stopChan:
+			return false
+		}
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), "", nil
 	}
-
 	return nil, responseText, usage
 }

@ -136,7 +149,7 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 		return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}

-	if textResponse.Usage.TotalTokens == 0 || (textResponse.Usage.PromptTokens == 0 && textResponse.Usage.CompletionTokens == 0) {
+	if textResponse.Usage.TotalTokens == 0 {
 		completionTokens := 0
 		for _, choice := range textResponse.Choices {
 			completionTokens += CountTokenText(choice.Message.StringContent(), modelName)
--- a/relay/adaptor/openai/token.go
+++ b/relay/adaptor/openai/token.go
@ -97,11 +97,7 @@ func CountTokenMessages(messages []model.Message, model string) int {
 				m := it.(map[string]any)
 				switch m["type"] {
 				case "text":
-					if textValue, ok := m["text"]; ok {
-						if textString, ok := textValue.(string); ok {
-							tokenNum += getTokenNum(tokenEncoder, textString)
-						}
-					}
+					tokenNum += getTokenNum(tokenEncoder, m["text"].(string))
 				case "image_url":
 					imageUrl, ok := m["image_url"].(map[string]any)
 					if ok {
@ -110,7 +106,7 @@ func CountTokenMessages(messages []model.Message, model string) int {
 						if imageUrl["detail"] != nil {
 							detail = imageUrl["detail"].(string)
 						}
-						imageTokens, err := countImageTokens(url, detail, model)
+						imageTokens, err := countImageTokens(url, detail)
 						if err != nil {
 							logger.SysError("error counting image tokens: " + err.Error())
 						} else {
@ -134,15 +130,11 @@ const (
 	lowDetailCost         = 85
 	highDetailCostPerTile = 170
 	additionalCost        = 85
-	// gpt-4o-mini cost higher than other model
-	gpt4oMiniLowDetailCost  = 2833
-	gpt4oMiniHighDetailCost = 5667
-	gpt4oMiniAdditionalCost = 2833
 )

 // https://platform.openai.com/docs/guides/vision/calculating-costs
 // https://github.com/openai/openai-cookbook/blob/05e3f9be4c7a2ae7ecf029a7c32065b024730ebe/examples/How_to_count_tokens_with_tiktoken.ipynb
-func countImageTokens(url string, detail string, model string) (_ int, err error) {
+func countImageTokens(url string, detail string) (_ int, err error) {
 	var fetchSize = true
 	var width, height int
 	// Reference: https://platform.openai.com/docs/guides/vision/low-or-high-fidelity-image-understanding
@ -176,9 +168,6 @@ func countImageTokens(url string, detail string, model string) (_ int, err error
 	}
 	switch detail {
 	case "low":
-		if strings.HasPrefix(model, "gpt-4o-mini") {
-			return gpt4oMiniLowDetailCost, nil
-		}
 		return lowDetailCost, nil
 	case "high":
 		if fetchSize {
@ -198,9 +187,6 @@ func countImageTokens(url string, detail string, model string) (_ int, err error
 			height = int(float64(height) * ratio)
 		}
 		numSquares := int(math.Ceil(float64(width)/512) * math.Ceil(float64(height)/512))
-		if strings.HasPrefix(model, "gpt-4o-mini") {
-			return numSquares*gpt4oMiniHighDetailCost + gpt4oMiniAdditionalCost, nil
-		}
 		result := numSquares*highDetailCostPerTile + additionalCost
 		return result, nil
 	default:
--- a/relay/adaptor/palm/palm.go
+++ b/relay/adaptor/palm/palm.go
@ -3,10 +3,6 @@ package palm
 import (
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/helper"
@ -15,6 +11,8 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
 )

 // https://developers.generativeai.google/api/rest/generativelanguage/models/generateMessage#request-body
@ -79,51 +77,58 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 	responseText := ""
 	responseId := fmt.Sprintf("chatcmpl-%s", random.GetUUID())
 	createdTime := helper.GetTimestamp()
-
-	common.SetEventStreamHeaders(c)
-
-	responseBody, err := io.ReadAll(resp.Body)
-	if err != nil {
-		logger.SysError("error reading stream response: " + err.Error())
-		err := resp.Body.Close()
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		responseBody, err := io.ReadAll(resp.Body)
 		if err != nil {
-			return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), ""
+			logger.SysError("error reading stream response: " + err.Error())
+			stopChan <- true
+			return
 		}
-		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), ""
-	}
-
-	err = resp.Body.Close()
+		err = resp.Body.Close()
+		if err != nil {
+			logger.SysError("error closing stream response: " + err.Error())
+			stopChan <- true
+			return
+		}
+		var palmResponse ChatResponse
+		err = json.Unmarshal(responseBody, &palmResponse)
+		if err != nil {
+			logger.SysError("error unmarshalling stream response: " + err.Error())
+			stopChan <- true
+			return
+		}
+		fullTextResponse := streamResponsePaLM2OpenAI(&palmResponse)
+		fullTextResponse.Id = responseId
+		fullTextResponse.Created = createdTime
+		if len(palmResponse.Candidates) > 0 {
+			responseText = palmResponse.Candidates[0].Content
+		}
+		jsonResponse, err := json.Marshal(fullTextResponse)
+		if err != nil {
+			logger.SysError("error marshalling stream response: " + err.Error())
+			stopChan <- true
+			return
+		}
+		dataChan <- string(jsonResponse)
+		stopChan <- true
+	}()
+	common.SetEventStreamHeaders(c)
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			c.Render(-1, common.CustomEvent{Data: "data: " + data})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
+		}
+	})
+	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), ""
 	}
-
-	var palmResponse ChatResponse
-	err = json.Unmarshal(responseBody, &palmResponse)
-	if err != nil {
-		logger.SysError("error unmarshalling stream response: " + err.Error())
-		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), ""
-	}
-
-	fullTextResponse := streamResponsePaLM2OpenAI(&palmResponse)
-	fullTextResponse.Id = responseId
-	fullTextResponse.Created = createdTime
-	if len(palmResponse.Candidates) > 0 {
-		responseText = palmResponse.Candidates[0].Content
-	}
-
-	jsonResponse, err := json.Marshal(fullTextResponse)
-	if err != nil {
-		logger.SysError("error marshalling stream response: " + err.Error())
-		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), ""
-	}
-
-	err = render.ObjectData(c, string(jsonResponse))
-	if err != nil {
-		logger.SysError(err.Error())
-	}
-
-	render.Done(c)
-
 	return nil, responseText
 }

--- a/relay/adaptor/proxy/adaptor.go
+++ b/relay/adaptor/proxy/adaptor.go
@ -1,89 +0,0 @@
-package proxy
-
-import (
-	"fmt"
-	"io"
-	"net/http"
-	"strings"
-
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/relay/adaptor"
-	channelhelper "github.com/songquanpeng/one-api/relay/adaptor"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-)
-
-var _ adaptor.Adaptor = new(Adaptor)
-
-const channelName = "proxy"
-
-type Adaptor struct{}
-
-func (a *Adaptor) Init(meta *meta.Meta) {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	return nil, errors.New("notimplement")
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	for k, v := range resp.Header {
-		for _, vv := range v {
-			c.Writer.Header().Set(k, vv)
-		}
-	}
-
-	c.Writer.WriteHeader(resp.StatusCode)
-	if _, gerr := io.Copy(c.Writer, resp.Body); gerr != nil {
-		return nil, &relaymodel.ErrorWithStatusCode{
-			StatusCode: http.StatusInternalServerError,
-			Error: relaymodel.Error{
-				Message: gerr.Error(),
-			},
-		}
-	}
-
-	return nil, nil
-}
-
-func (a *Adaptor) GetModelList() (models []string) {
-	return nil
-}
-
-func (a *Adaptor) GetChannelName() string {
-	return channelName
-}
-
-// GetRequestURL remove static prefix, and return the real request url to the upstream service
-func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	prefix := fmt.Sprintf("/v1/oneapi/proxy/%d", meta.ChannelId)
-	return meta.BaseURL + strings.TrimPrefix(meta.RequestURLPath, prefix), nil
-
-}
-
-func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
-	for k, v := range c.Request.Header {
-		req.Header.Set(k, v[0])
-	}
-
-	// remove unnecessary headers
-	req.Header.Del("Host")
-	req.Header.Del("Content-Length")
-	req.Header.Del("Accept-Encoding")
-	req.Header.Del("Connection")
-
-	// set authorization header
-	req.Header.Set("Authorization", meta.APIKey)
-
-	return nil
-}
-
-func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
-	return nil, errors.Errorf("not implement")
-}
-
-func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
-	return channelhelper.DoRequestHelper(a, c, meta, requestBody)
-}
--- a/relay/adaptor/siliconflow/constants.go
+++ b/relay/adaptor/siliconflow/constants.go
@ -1,36 +0,0 @@
-package siliconflow
-
-// https://docs.siliconflow.cn/docs/getting-started
-
-var ModelList = []string{
-	"deepseek-ai/deepseek-llm-67b-chat",
-	"Qwen/Qwen1.5-14B-Chat",
-	"Qwen/Qwen1.5-7B-Chat",
-	"Qwen/Qwen1.5-110B-Chat",
-	"Qwen/Qwen1.5-32B-Chat",
-	"01-ai/Yi-1.5-6B-Chat",
-	"01-ai/Yi-1.5-9B-Chat-16K",
-	"01-ai/Yi-1.5-34B-Chat-16K",
-	"THUDM/chatglm3-6b",
-	"deepseek-ai/DeepSeek-V2-Chat",
-	"THUDM/glm-4-9b-chat",
-	"Qwen/Qwen2-72B-Instruct",
-	"Qwen/Qwen2-7B-Instruct",
-	"Qwen/Qwen2-57B-A14B-Instruct",
-	"deepseek-ai/DeepSeek-Coder-V2-Instruct",
-	"Qwen/Qwen2-1.5B-Instruct",
-	"internlm/internlm2_5-7b-chat",
-	"BAAI/bge-large-en-v1.5",
-	"BAAI/bge-large-zh-v1.5",
-	"Pro/Qwen/Qwen2-7B-Instruct",
-	"Pro/Qwen/Qwen2-1.5B-Instruct",
-	"Pro/Qwen/Qwen1.5-7B-Chat",
-	"Pro/THUDM/glm-4-9b-chat",
-	"Pro/THUDM/chatglm3-6b",
-	"Pro/01-ai/Yi-1.5-9B-Chat-16K",
-	"Pro/01-ai/Yi-1.5-6B-Chat",
-	"Pro/google/gemma-2-9b-it",
-	"Pro/internlm/internlm2_5-7b-chat",
-	"Pro/meta-llama/Meta-Llama-3-8B-Instruct",
-	"Pro/mistralai/Mistral-7B-Instruct-v0.2",
-}
--- a/relay/adaptor/stepfun/constants.go
+++ b/relay/adaptor/stepfun/constants.go
@ -1,13 +1,7 @@
 package stepfun

 var ModelList = []string{
-	"step-1-8k",
 	"step-1-32k",
-	"step-1-128k",
-	"step-1-256k",
-	"step-1-flash",
-	"step-2-16k",
-	"step-1v-8k",
 	"step-1v-32k",
-	"step-1x-medium",
+	"step-1-200k",
 }
--- a/relay/adaptor/tencent/constants.go
+++ b/relay/adaptor/tencent/constants.go
@ -5,5 +5,4 @@ var ModelList = []string{
 	"hunyuan-standard",
 	"hunyuan-standard-256K",
 	"hunyuan-pro",
-	"hunyuan-vision",
 }
--- a/relay/adaptor/tencent/main.go
+++ b/relay/adaptor/tencent/main.go
@ -8,13 +8,6 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strconv"
-	"strings"
-	"time"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/conv"
@ -24,6 +17,11 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+	"time"
 )

 func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
@ -89,46 +87,64 @@ func streamResponseTencent2OpenAI(TencentResponse *ChatResponse) *openai.ChatCom
 func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, string) {
 	var responseText string
 	scanner := bufio.NewScanner(resp.Body)
-	scanner.Split(bufio.ScanLines)
-
+	scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
+		if atEOF && len(data) == 0 {
+			return 0, nil, nil
+		}
+		if i := strings.Index(string(data), "\n"); i >= 0 {
+			return i + 1, data[0:i], nil
+		}
+		if atEOF {
+			return len(data), data, nil
+		}
+		return 0, nil, nil
+	})
+	dataChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			if len(data) < 5 { // ignore blank line or wrong format
+				continue
+			}
+			if data[:5] != "data:" {
+				continue
+			}
+			data = data[5:]
+			dataChan <- data
+		}
+		stopChan <- true
+	}()
 	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		if len(data) < 5 || !strings.HasPrefix(data, "data:") {
-			continue
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			var TencentResponse ChatResponse
+			err := json.Unmarshal([]byte(data), &TencentResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response := streamResponseTencent2OpenAI(&TencentResponse)
+			if len(response.Choices) != 0 {
+				responseText += conv.AsString(response.Choices[0].Delta.Content)
+			}
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
 		}
-		data = strings.TrimPrefix(data, "data:")
-
-		var tencentResponse ChatResponse
-		err := json.Unmarshal([]byte(data), &tencentResponse)
-		if err != nil {
-			logger.SysError("error unmarshalling stream response: " + err.Error())
-			continue
-		}
-
-		response := streamResponseTencent2OpenAI(&tencentResponse)
-		if len(response.Choices) != 0 {
-			responseText += conv.AsString(response.Choices[0].Delta.Content)
-		}
-
-		err = render.ObjectData(c, response)
-		if err != nil {
-			logger.SysError(err.Error())
-		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), ""
 	}
-
 	return nil, responseText
 }

--- a/relay/adaptor/vertexai/adaptor.go
+++ b/relay/adaptor/vertexai/adaptor.go
@ -1,117 +0,0 @@
-package vertexai
-
-import (
-	"errors"
-	"fmt"
-	"io"
-	"net/http"
-	"strings"
-
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/relay/adaptor"
-	channelhelper "github.com/songquanpeng/one-api/relay/adaptor"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-)
-
-var _ adaptor.Adaptor = new(Adaptor)
-
-const channelName = "vertexai"
-
-type Adaptor struct{}
-
-func (a *Adaptor) Init(meta *meta.Meta) {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	adaptor := GetAdaptor(request.Model)
-	if adaptor == nil {
-		return nil, errors.New("adaptor not found")
-	}
-
-	return adaptor.ConvertRequest(c, relayMode, request)
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	adaptor := GetAdaptor(meta.ActualModelName)
-	if adaptor == nil {
-		return nil, &relaymodel.ErrorWithStatusCode{
-			StatusCode: http.StatusInternalServerError,
-			Error: relaymodel.Error{
-				Message: "adaptor not found",
-			},
-		}
-	}
-	return adaptor.DoResponse(c, resp, meta)
-}
-
-func (a *Adaptor) GetModelList() (models []string) {
-	models = modelList
-	return
-}
-
-func (a *Adaptor) GetChannelName() string {
-	return channelName
-}
-
-func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	suffix := ""
-	if strings.HasPrefix(meta.ActualModelName, "gemini") {
-		if meta.IsStream {
-			suffix = "streamGenerateContent?alt=sse"
-		} else {
-			suffix = "generateContent"
-		}
-	} else {
-		if meta.IsStream {
-			suffix = "streamRawPredict?alt=sse"
-		} else {
-			suffix = "rawPredict"
-		}
-	}
-
-	if meta.BaseURL != "" {
-		return fmt.Sprintf(
-			"%s/v1/projects/%s/locations/%s/publishers/google/models/%s:%s",
-			meta.BaseURL,
-			meta.Config.VertexAIProjectID,
-			meta.Config.Region,
-			meta.ActualModelName,
-			suffix,
-		), nil
-	}
-	return fmt.Sprintf(
-		"https://%s-aiplatform.googleapis.com/v1/projects/%s/locations/%s/publishers/google/models/%s:%s",
-		meta.Config.Region,
-		meta.Config.VertexAIProjectID,
-		meta.Config.Region,
-		meta.ActualModelName,
-		suffix,
-	), nil
-}
-
-func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
-	adaptor.SetupCommonRequestHeader(c, req, meta)
-	token, err := getToken(c, meta.ChannelId, meta.Config.VertexAIADC)
-	if err != nil {
-		return err
-	}
-	req.Header.Set("Authorization", "Bearer "+token)
-	return nil
-}
-
-func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-	return request, nil
-}
-
-func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
-	return channelhelper.DoRequestHelper(a, c, meta, requestBody)
-}
--- a/relay/adaptor/vertexai/claude/adapter.go
+++ b/relay/adaptor/vertexai/claude/adapter.go
@ -1,55 +0,0 @@
-package vertexai
-
-import (
-	"net/http"
-
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
-
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-var ModelList = []string{
-	"claude-3-haiku@20240307", "claude-3-opus@20240229", "claude-3-5-sonnet@20240620", "claude-3-sonnet@20240229",
-}
-
-const anthropicVersion = "vertex-2023-10-16"
-
-type Adaptor struct {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	claudeReq := anthropic.ConvertRequest(*request)
-	req := Request{
-		AnthropicVersion: anthropicVersion,
-		// Model:            claudeReq.Model,
-		Messages:    claudeReq.Messages,
-		System:      claudeReq.System,
-		MaxTokens:   claudeReq.MaxTokens,
-		Temperature: claudeReq.Temperature,
-		TopP:        claudeReq.TopP,
-		TopK:        claudeReq.TopK,
-		Stream:      claudeReq.Stream,
-		Tools:       claudeReq.Tools,
-	}
-
-	c.Set(ctxkey.RequestModel, request.Model)
-	c.Set(ctxkey.ConvertedRequest, req)
-	return req, nil
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	if meta.IsStream {
-		err, usage = anthropic.StreamHandler(c, resp)
-	} else {
-		err, usage = anthropic.Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
-	}
-	return
-}
--- a/relay/adaptor/vertexai/claude/model.go
+++ b/relay/adaptor/vertexai/claude/model.go
@ -1,19 +0,0 @@
-package vertexai
-
-import "github.com/songquanpeng/one-api/relay/adaptor/anthropic"
-
-type Request struct {
-	// AnthropicVersion must be "vertex-2023-10-16"
-	AnthropicVersion string `json:"anthropic_version"`
-	// Model            string              `json:"model"`
-	Messages      []anthropic.Message `json:"messages"`
-	System        string              `json:"system,omitempty"`
-	MaxTokens     int                 `json:"max_tokens,omitempty"`
-	StopSequences []string            `json:"stop_sequences,omitempty"`
-	Stream        bool                `json:"stream,omitempty"`
-	Temperature   float64             `json:"temperature,omitempty"`
-	TopP          float64             `json:"top_p,omitempty"`
-	TopK          int                 `json:"top_k,omitempty"`
-	Tools         []anthropic.Tool    `json:"tools,omitempty"`
-	ToolChoice    any                 `json:"tool_choice,omitempty"`
-}
--- a/relay/adaptor/vertexai/gemini/adapter.go
+++ b/relay/adaptor/vertexai/gemini/adapter.go
@ -1,49 +0,0 @@
-package vertexai
-
-import (
-	"net/http"
-
-	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"github.com/songquanpeng/one-api/relay/adaptor/gemini"
-	"github.com/songquanpeng/one-api/relay/adaptor/openai"
-	"github.com/songquanpeng/one-api/relay/relaymode"
-
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-var ModelList = []string{
-	"gemini-1.5-pro-001", "gemini-1.5-flash-001", "gemini-pro", "gemini-pro-vision",
-}
-
-type Adaptor struct {
-}
-
-func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
-	if request == nil {
-		return nil, errors.New("request is nil")
-	}
-
-	geminiRequest := gemini.ConvertRequest(*request)
-	c.Set(ctxkey.RequestModel, request.Model)
-	c.Set(ctxkey.ConvertedRequest, geminiRequest)
-	return geminiRequest, nil
-}
-
-func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
-	if meta.IsStream {
-		var responseText string
-		err, responseText = gemini.StreamHandler(c, resp)
-		usage = openai.ResponseText2Usage(responseText, meta.ActualModelName, meta.PromptTokens)
-	} else {
-		switch meta.Mode {
-		case relaymode.Embeddings:
-			err, usage = gemini.EmbeddingHandler(c, resp)
-		default:
-			err, usage = gemini.Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
-		}
-	}
-	return
-}
--- a/relay/adaptor/vertexai/registry.go
+++ b/relay/adaptor/vertexai/registry.go
@ -1,50 +0,0 @@
-package vertexai
-
-import (
-	"net/http"
-
-	"github.com/gin-gonic/gin"
-	claude "github.com/songquanpeng/one-api/relay/adaptor/vertexai/claude"
-	gemini "github.com/songquanpeng/one-api/relay/adaptor/vertexai/gemini"
-	"github.com/songquanpeng/one-api/relay/meta"
-	"github.com/songquanpeng/one-api/relay/model"
-)
-
-type VertexAIModelType int
-
-const (
-	VerterAIClaude VertexAIModelType = iota + 1
-	VerterAIGemini
-)
-
-var modelMapping = map[string]VertexAIModelType{}
-var modelList = []string{}
-
-func init() {
-	modelList = append(modelList, claude.ModelList...)
-	for _, model := range claude.ModelList {
-		modelMapping[model] = VerterAIClaude
-	}
-
-	modelList = append(modelList, gemini.ModelList...)
-	for _, model := range gemini.ModelList {
-		modelMapping[model] = VerterAIGemini
-	}
-}
-
-type innerAIAdapter interface {
-	ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error)
-	DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode)
-}
-
-func GetAdaptor(model string) innerAIAdapter {
-	adaptorType := modelMapping[model]
-	switch adaptorType {
-	case VerterAIClaude:
-		return &claude.Adaptor{}
-	case VerterAIGemini:
-		return &gemini.Adaptor{}
-	default:
-		return nil
-	}
-}
--- a/relay/adaptor/vertexai/token.go
+++ b/relay/adaptor/vertexai/token.go
@ -1,62 +0,0 @@
-package vertexai
-
-import (
-	"context"
-	"encoding/json"
-	"fmt"
-	"time"
-
-	credentials "cloud.google.com/go/iam/credentials/apiv1"
-	"cloud.google.com/go/iam/credentials/apiv1/credentialspb"
-	"github.com/patrickmn/go-cache"
-	"google.golang.org/api/option"
-)
-
-type ApplicationDefaultCredentials struct {
-	Type                    string `json:"type"`
-	ProjectID               string `json:"project_id"`
-	PrivateKeyID            string `json:"private_key_id"`
-	PrivateKey              string `json:"private_key"`
-	ClientEmail             string `json:"client_email"`
-	ClientID                string `json:"client_id"`
-	AuthURI                 string `json:"auth_uri"`
-	TokenURI                string `json:"token_uri"`
-	AuthProviderX509CertURL string `json:"auth_provider_x509_cert_url"`
-	ClientX509CertURL       string `json:"client_x509_cert_url"`
-	UniverseDomain          string `json:"universe_domain"`
-}
-
-var Cache = cache.New(50*time.Minute, 55*time.Minute)
-
-const defaultScope = "https://www.googleapis.com/auth/cloud-platform"
-
-func getToken(ctx context.Context, channelId int, adcJson string) (string, error) {
-	cacheKey := fmt.Sprintf("vertexai-token-%d", channelId)
-	if token, found := Cache.Get(cacheKey); found {
-		return token.(string), nil
-	}
-	adc := &ApplicationDefaultCredentials{}
-	if err := json.Unmarshal([]byte(adcJson), adc); err != nil {
-		return "", fmt.Errorf("Failed to decode credentials file: %w", err)
-	}
-
-	c, err := credentials.NewIamCredentialsClient(ctx, option.WithCredentialsJSON([]byte(adcJson)))
-	if err != nil {
-		return "", fmt.Errorf("Failed to create client: %w", err)
-	}
-	defer c.Close()
-
-	req := &credentialspb.GenerateAccessTokenRequest{
-		// See https://pkg.go.dev/cloud.google.com/go/iam/credentials/apiv1/credentialspb#GenerateAccessTokenRequest.
-		Name:  fmt.Sprintf("projects/-/serviceAccounts/%s", adc.ClientEmail),
-		Scope: []string{defaultScope},
-	}
-	resp, err := c.GenerateAccessToken(ctx, req)
-	if err != nil {
-		return "", fmt.Errorf("Failed to generate access token: %w", err)
-	}
-	_ = resp
-
-	Cache.Set(cacheKey, resp.AccessToken, cache.DefaultExpiration)
-	return resp.AccessToken, nil
-}
--- a/relay/adaptor/xunfei/constants.go
+++ b/relay/adaptor/xunfei/constants.go
@ -5,7 +5,5 @@ var ModelList = []string{
 	"SparkDesk-v1.1",
 	"SparkDesk-v2.1",
 	"SparkDesk-v3.1",
-	"SparkDesk-v3.1-128K",
 	"SparkDesk-v3.5",
-	"SparkDesk-v4.0",
 }
--- a/relay/adaptor/xunfei/main.go
+++ b/relay/adaptor/xunfei/main.go
@ -44,7 +44,7 @@ func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string
 	xunfeiRequest.Parameter.Chat.MaxTokens = request.MaxTokens
 	xunfeiRequest.Payload.Message.Text = messages

-	if strings.HasPrefix(domain, "generalv3") || domain == "4.0Ultra" {
+	if strings.HasPrefix(domain, "generalv3") {
 		functions := make([]model.Function, len(request.Tools))
 		for i, tool := range request.Tools {
 			functions[i] = tool.Function
@ -272,9 +272,9 @@ func xunfeiMakeRequest(textRequest model.GeneralOpenAIRequest, domain, authUrl,
 }

 func parseAPIVersionByModelName(modelName string) string {
-	index := strings.IndexAny(modelName, "-")
-	if index != -1 {
-		return modelName[index+1:]
+	parts := strings.Split(modelName, "-")
+	if len(parts) == 2 {
+		return parts[1]
 	}
 	return ""
 }
@ -288,25 +288,14 @@ func apiVersion2domain(apiVersion string) string {
 		return "generalv2"
 	case "v3.1":
 		return "generalv3"
-	case "v3.1-128K":
-		return "pro-128k"
 	case "v3.5":
 		return "generalv3.5"
-	case "v4.0":
-		return "4.0Ultra"
 	}
 	return "general" + apiVersion
 }

 func getXunfeiAuthUrl(apiVersion string, apiKey string, apiSecret string) (string, string) {
-	var authUrl string
 	domain := apiVersion2domain(apiVersion)
-	switch apiVersion {
-	case "v3.1-128K":
-		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/pro-128k", apiVersion), apiKey, apiSecret)
-		break
-	default:
-		authUrl = buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
-	}
+	authUrl := buildXunfeiAuthUrl(fmt.Sprintf("wss://spark-api.xf-yun.com/%s/chat", apiVersion), apiKey, apiSecret)
 	return domain, authUrl
 }
--- a/relay/adaptor/zhipu/main.go
+++ b/relay/adaptor/zhipu/main.go
@ -3,13 +3,6 @@ package zhipu
 import (
 	"bufio"
 	"encoding/json"
-	"github.com/songquanpeng/one-api/common/render"
-	"io"
-	"net/http"
-	"strings"
-	"sync"
-	"time"
-
 	"github.com/gin-gonic/gin"
 	"github.com/golang-jwt/jwt"
 	"github.com/songquanpeng/one-api/common"
@ -18,6 +11,11 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/constant"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+	"strings"
+	"sync"
+	"time"
 )

 // https://open.bigmodel.cn/doc/api#chatglm_std
@ -157,55 +155,66 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-
-	common.SetEventStreamHeaders(c)
-
-	for scanner.Scan() {
-		data := scanner.Text()
-		lines := strings.Split(data, "\n")
-		for i, line := range lines {
-			if len(line) < 5 {
-				continue
-			}
-			if strings.HasPrefix(line, "data:") {
-				dataSegment := line[5:]
-				if i != len(lines)-1 {
-					dataSegment += "\n"
-				}
-				response := streamResponseZhipu2OpenAI(dataSegment)
-				err := render.ObjectData(c, response)
-				if err != nil {
-					logger.SysError("error marshalling stream response: " + err.Error())
-				}
-			} else if strings.HasPrefix(line, "meta:") {
-				metaSegment := line[5:]
-				var zhipuResponse StreamMetaResponse
-				err := json.Unmarshal([]byte(metaSegment), &zhipuResponse)
-				if err != nil {
-					logger.SysError("error unmarshalling stream response: " + err.Error())
+	dataChan := make(chan string)
+	metaChan := make(chan string)
+	stopChan := make(chan bool)
+	go func() {
+		for scanner.Scan() {
+			data := scanner.Text()
+			lines := strings.Split(data, "\n")
+			for i, line := range lines {
+				if len(line) < 5 {
 					continue
 				}
-				response, zhipuUsage := streamMetaResponseZhipu2OpenAI(&zhipuResponse)
-				err = render.ObjectData(c, response)
-				if err != nil {
-					logger.SysError("error marshalling stream response: " + err.Error())
+				if line[:5] == "data:" {
+					dataChan <- line[5:]
+					if i != len(lines)-1 {
+						dataChan <- "\n"
+					}
+				} else if line[:5] == "meta:" {
+					metaChan <- line[5:]
 				}
-				usage = zhipuUsage
 			}
 		}
-	}
-
-	if err := scanner.Err(); err != nil {
-		logger.SysError("error reading stream: " + err.Error())
-	}
-
-	render.Done(c)
-
+		stopChan <- true
+	}()
+	common.SetEventStreamHeaders(c)
+	c.Stream(func(w io.Writer) bool {
+		select {
+		case data := <-dataChan:
+			response := streamResponseZhipu2OpenAI(data)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case data := <-metaChan:
+			var zhipuResponse StreamMetaResponse
+			err := json.Unmarshal([]byte(data), &zhipuResponse)
+			if err != nil {
+				logger.SysError("error unmarshalling stream response: " + err.Error())
+				return true
+			}
+			response, zhipuUsage := streamMetaResponseZhipu2OpenAI(&zhipuResponse)
+			jsonResponse, err := json.Marshal(response)
+			if err != nil {
+				logger.SysError("error marshalling stream response: " + err.Error())
+				return true
+			}
+			usage = zhipuUsage
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
+			return true
+		case <-stopChan:
+			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+			return false
+		}
+	})
 	err := resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
-
 	return nil, usage
 }

--- a/relay/apitype/define.go
+++ b/relay/apitype/define.go
@ -17,8 +17,6 @@ const (
 	Cohere
 	Cloudflare
 	DeepL
-	VertexAI
-	Proxy

 	Dummy // this one is only for count, do not add any channel after this
 )
--- a/relay/billing/ratio/image.go
+++ b/relay/billing/ratio/image.go
@ -30,14 +30,6 @@ var ImageSizeRatios = map[string]map[string]float64{
 		"720x1280":  1,
 		"1280x720":  1,
 	},
-	"step-1x-medium": {
-		"256x256":   1,
-		"512x512":   1,
-		"768x768":   1,
-		"1024x1024": 1,
-		"1280x800":  1,
-		"800x1280":  1,
-	},
 }

 var ImageGenerationAmounts = map[string][2]int{
@ -47,7 +39,6 @@ var ImageGenerationAmounts = map[string][2]int{
 	"ali-stable-diffusion-v1.5": {1, 4}, // Ali
 	"wanx-v1":                   {1, 4}, // Ali
 	"cogview-3":                 {1, 1},
-	"step-1x-medium":            {1, 1},
 }

 var ImagePromptLengthLimitations = map[string]int{
@ -57,7 +48,6 @@ var ImagePromptLengthLimitations = map[string]int{
 	"ali-stable-diffusion-v1.5": 4000,
 	"wanx-v1":                   4000,
 	"cogview-3":                 833,
-	"step-1x-medium":            4000,
 }

 var ImageOriginModelName = map[string]string{
--- a/relay/billing/ratio/model.go
+++ b/relay/billing/ratio/model.go
@ -2,7 +2,6 @@ package ratio

 import (
 	"encoding/json"
-	"fmt"
 	"strings"

 	"github.com/songquanpeng/one-api/common/logger"
@ -28,19 +27,15 @@ var ModelRatio = map[string]float64{
 	"gpt-4-32k":               30,
 	"gpt-4-32k-0314":          30,
 	"gpt-4-32k-0613":          30,
-	"gpt-4-1106-preview":      5,     // $0.01 / 1K tokens
-	"gpt-4-0125-preview":      5,     // $0.01 / 1K tokens
-	"gpt-4-turbo-preview":     5,     // $0.01 / 1K tokens
-	"gpt-4-turbo":             5,     // $0.01 / 1K tokens
-	"gpt-4-turbo-2024-04-09":  5,     // $0.01 / 1K tokens
-	"gpt-4o":                  2.5,   // $0.005 / 1K tokens
-	"chatgpt-4o-latest":       2.5,   // $0.005 / 1K tokens
-	"gpt-4o-2024-05-13":       2.5,   // $0.005 / 1K tokens
-	"gpt-4o-2024-08-06":       1.25,  // $0.0025 / 1K tokens
-	"gpt-4o-mini":             0.075, // $0.00015 / 1K tokens
-	"gpt-4o-mini-2024-07-18":  0.075, // $0.00015 / 1K tokens
-	"gpt-4-vision-preview":    5,     // $0.01 / 1K tokens
-	"gpt-3.5-turbo":           0.25,  // $0.0005 / 1K tokens
+	"gpt-4-1106-preview":      5,    // $0.01 / 1K tokens
+	"gpt-4-0125-preview":      5,    // $0.01 / 1K tokens
+	"gpt-4-turbo-preview":     5,    // $0.01 / 1K tokens
+	"gpt-4-turbo":             5,    // $0.01 / 1K tokens
+	"gpt-4-turbo-2024-04-09":  5,    // $0.01 / 1K tokens
+	"gpt-4o":                  2.5,  // $0.005 / 1K tokens
+	"gpt-4o-2024-05-13":       2.5,  // $0.005 / 1K tokens
+	"gpt-4-vision-preview":    5,    // $0.01 / 1K tokens
+	"gpt-3.5-turbo":           0.25, // $0.0005 / 1K tokens
 	"gpt-3.5-turbo-0301":      0.75,
 	"gpt-3.5-turbo-0613":      0.75,
 	"gpt-3.5-turbo-16k":       1.5, // $0.003 / 1K tokens
@ -75,13 +70,12 @@ var ModelRatio = map[string]float64{
 	"dall-e-2":                0.02 * USD, // $0.016 - $0.020 / image
 	"dall-e-3":                0.04 * USD, // $0.040 - $0.120 / image
 	// https://www.anthropic.com/api#pricing
-	"claude-instant-1.2":         0.8 / 1000 * USD,
-	"claude-2.0":                 8.0 / 1000 * USD,
-	"claude-2.1":                 8.0 / 1000 * USD,
-	"claude-3-haiku-20240307":    0.25 / 1000 * USD,
-	"claude-3-sonnet-20240229":   3.0 / 1000 * USD,
-	"claude-3-5-sonnet-20240620": 3.0 / 1000 * USD,
-	"claude-3-opus-20240229":     15.0 / 1000 * USD,
+	"claude-instant-1.2":       0.8 / 1000 * USD,
+	"claude-2.0":               8.0 / 1000 * USD,
+	"claude-2.1":               8.0 / 1000 * USD,
+	"claude-3-haiku-20240307":  0.25 / 1000 * USD,
+	"claude-3-sonnet-20240229": 3.0 / 1000 * USD,
+	"claude-3-opus-20240229":   15.0 / 1000 * USD,
 	// https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4akp7
 	"ERNIE-4.0-8K":       0.120 * RMB,
 	"ERNIE-3.5-8K":       0.012 * RMB,
@ -100,11 +94,12 @@ var ModelRatio = map[string]float64{
 	"bge-large-en":       0.002 * RMB,
 	"tao-8k":             0.002 * RMB,
 	// https://ai.google.dev/pricing
-	"gemini-pro":       1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
-	"gemini-1.0-pro":   1,
-	"gemini-1.5-flash": 1,
-	"gemini-1.5-pro":   1,
-	"aqa":              1,
+	"PaLM-2":                    1,
+	"gemini-pro":                1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-pro-vision":         1, // $0.00025 / 1k characters -> $0.001 / 1k tokens
+	"gemini-1.0-pro-vision-001": 1,
+	"gemini-1.0-pro-001":        1,
+	"gemini-1.5-pro":            1,
 	// https://open.bigmodel.cn/pricing
 	"glm-4":         0.1 * RMB,
 	"glm-4v":        0.1 * RMB,
@ -128,9 +123,7 @@ var ModelRatio = map[string]float64{
 	"SparkDesk-v1.1":            1.2858, // ￥0.018 / 1k tokens
 	"SparkDesk-v2.1":            1.2858, // ￥0.018 / 1k tokens
 	"SparkDesk-v3.1":            1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v3.1-128K":       1.2858, // ￥0.018 / 1k tokens
 	"SparkDesk-v3.5":            1.2858, // ￥0.018 / 1k tokens
-	"SparkDesk-v4.0":            1.2858, // ￥0.018 / 1k tokens
 	"360GPT_S2_V9":              0.8572, // ¥0.012 / 1k tokens
 	"embedding-bert-512-v1":     0.0715, // ¥0.001 / 1k tokens
 	"embedding_s1_v1":           0.0715, // ¥0.001 / 1k tokens
@ -160,32 +153,20 @@ var ModelRatio = map[string]float64{
 	"mistral-large-latest":  8.0 / 1000 * USD,
 	"mistral-embed":         0.1 / 1000 * USD,
 	// https://wow.groq.com/#:~:text=inquiries%C2%A0here.-,Model,-Current%20Speed
-	"gemma-7b-it":                           0.07 / 1000000 * USD,
-	"mixtral-8x7b-32768":                    0.24 / 1000000 * USD,
-	"llama3-8b-8192":                        0.05 / 1000000 * USD,
-	"llama3-70b-8192":                       0.59 / 1000000 * USD,
-	"gemma2-9b-it":                          0.20 / 1000000 * USD,
-	"llama-3.1-405b-reasoning":              0.89 / 1000000 * USD,
-	"llama-3.1-70b-versatile":               0.59 / 1000000 * USD,
-	"llama-3.1-8b-instant":                  0.05 / 1000000 * USD,
-	"llama3-groq-70b-8192-tool-use-preview": 0.89 / 1000000 * USD,
-	"llama3-groq-8b-8192-tool-use-preview":  0.19 / 1000000 * USD,
+	"llama3-70b-8192":    0.59 / 1000 * USD,
+	"mixtral-8x7b-32768": 0.27 / 1000 * USD,
+	"llama3-8b-8192":     0.05 / 1000 * USD,
+	"gemma-7b-it":        0.1 / 1000 * USD,
+	"llama2-70b-4096":    0.64 / 1000 * USD,
+	"llama2-7b-2048":     0.1 / 1000 * USD,
 	// https://platform.lingyiwanwu.com/docs#-计费单元
 	"yi-34b-chat-0205": 2.5 / 1000 * RMB,
 	"yi-34b-chat-200k": 12.0 / 1000 * RMB,
 	"yi-vl-plus":       6.0 / 1000 * RMB,
-	// https://platform.stepfun.com/docs/pricing/details
-	"step-1-8k":    0.005 / 1000 * RMB,
-	"step-1-32k":   0.015 / 1000 * RMB,
-	"step-1-128k":  0.040 / 1000 * RMB,
-	"step-1-256k":  0.095 / 1000 * RMB,
-	"step-1-flash": 0.001 / 1000 * RMB,
-	"step-2-16k":   0.038 / 1000 * RMB,
-	"step-1v-8k":   0.005 / 1000 * RMB,
-	"step-1v-32k":  0.015 / 1000 * RMB,
-	// aws llama3 https://aws.amazon.com/cn/bedrock/pricing/
-	"llama3-8b-8192(33)":  0.0003 / 0.002,  // $0.0003 / 1K tokens
-	"llama3-70b-8192(33)": 0.00265 / 0.002, // $0.00265 / 1K tokens
+	// stepfun todo
+	"step-1v-32k": 0.024 * RMB,
+	"step-1-32k":  0.024 * RMB,
+	"step-1-200k": 0.15 * RMB,
 	// https://cohere.com/pricing
 	"command":               0.5,
 	"command-nightly":       0.5,
@ -202,16 +183,10 @@ var ModelRatio = map[string]float64{
 	"deepl-ja": 25.0 / 1000 * USD,
 }

-var CompletionRatio = map[string]float64{
-	// aws llama3
-	"llama3-8b-8192(33)":  0.0006 / 0.0003,
-	"llama3-70b-8192(33)": 0.0035 / 0.00265,
-}
+var CompletionRatio = map[string]float64{}

-var (
-	DefaultModelRatio      map[string]float64
-	DefaultCompletionRatio map[string]float64
-)
+var DefaultModelRatio map[string]float64
+var DefaultCompletionRatio map[string]float64

 func init() {
 	DefaultModelRatio = make(map[string]float64)
@ -257,28 +232,22 @@ func UpdateModelRatioByJSONString(jsonStr string) error {
 	return json.Unmarshal([]byte(jsonStr), &ModelRatio)
 }

-func GetModelRatio(name string, channelType int) float64 {
+func GetModelRatio(name string) float64 {
 	if strings.HasPrefix(name, "qwen-") && strings.HasSuffix(name, "-internet") {
 		name = strings.TrimSuffix(name, "-internet")
 	}
 	if strings.HasPrefix(name, "command-") && strings.HasSuffix(name, "-internet") {
 		name = strings.TrimSuffix(name, "-internet")
 	}
-	model := fmt.Sprintf("%s(%d)", name, channelType)
-	if ratio, ok := ModelRatio[model]; ok {
-		return ratio
+	ratio, ok := ModelRatio[name]
+	if !ok {
+		ratio, ok = DefaultModelRatio[name]
 	}
-	if ratio, ok := DefaultModelRatio[model]; ok {
-		return ratio
+	if !ok {
+		logger.SysError("model ratio not found: " + name)
+		return 30
 	}
-	if ratio, ok := ModelRatio[name]; ok {
-		return ratio
-	}
-	if ratio, ok := DefaultModelRatio[name]; ok {
-		return ratio
-	}
-	logger.SysError("model ratio not found: " + name)
-	return 30
+	return ratio
 }

 func CompletionRatio2JSONString() string {
@ -294,17 +263,7 @@ func UpdateCompletionRatioByJSONString(jsonStr string) error {
 	return json.Unmarshal([]byte(jsonStr), &CompletionRatio)
 }

-func GetCompletionRatio(name string, channelType int) float64 {
-	if strings.HasPrefix(name, "qwen-") && strings.HasSuffix(name, "-internet") {
-		name = strings.TrimSuffix(name, "-internet")
-	}
-	model := fmt.Sprintf("%s(%d)", name, channelType)
-	if ratio, ok := CompletionRatio[model]; ok {
-		return ratio
-	}
-	if ratio, ok := DefaultCompletionRatio[model]; ok {
-		return ratio
-	}
+func GetCompletionRatio(name string) float64 {
 	if ratio, ok := CompletionRatio[name]; ok {
 		return ratio
 	}
@ -323,9 +282,6 @@ func GetCompletionRatio(name string, channelType int) float64 {
 		return 4.0 / 3.0
 	}
 	if strings.HasPrefix(name, "gpt-4") {
-		if strings.HasPrefix(name, "gpt-4o-mini") || name == "gpt-4o-2024-08-06" {
-			return 4
-		}
 		if strings.HasPrefix(name, "gpt-4-turbo") ||
 			strings.HasPrefix(name, "gpt-4o") ||
 			strings.HasSuffix(name, "preview") {
@ -333,9 +289,6 @@ func GetCompletionRatio(name string, channelType int) float64 {
 		}
 		return 2
 	}
-	if name == "chatgpt-4o-latest" {
-		return 3
-	}
 	if strings.HasPrefix(name, "claude-3") {
 		return 5
 	}
--- a/relay/channeltype/define.go
+++ b/relay/channeltype/define.go
@ -42,9 +42,5 @@ const (
 	DeepL
 	TogetherAI
 	Doubao
-	Novita
-	VertextAI
-	Proxy
-	SiliconFlow
 	Dummy
 )
--- a/relay/channeltype/helper.go
+++ b/relay/channeltype/helper.go
@ -35,10 +35,6 @@ func ToAPIType(channelType int) int {
 		apiType = apitype.Cloudflare
 	case DeepL:
 		apiType = apitype.DeepL
-	case VertextAI:
-		apiType = apitype.VertexAI
-	case Proxy:
-		apiType = apitype.Proxy
 	}

 	return apiType
--- a/relay/channeltype/url.go
+++ b/relay/channeltype/url.go
@ -42,10 +42,6 @@ var ChannelBaseURLs = []string{
 	"https://api-free.deepl.com",                // 38
 	"https://api.together.xyz",                  // 39
 	"https://ark.cn-beijing.volces.com",         // 40
-	"https://api.novita.ai/v3/openai",           // 41
-	"",                                          // 42
-	"",                                          // 43
-	"https://api.siliconflow.cn",                 // 44
 }

 func init() {
--- a/relay/controller/audio.go
+++ b/relay/controller/audio.go
@ -7,10 +7,6 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"io"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/client"
@ -25,6 +21,9 @@ import (
 	"github.com/songquanpeng/one-api/relay/meta"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
+	"io"
+	"net/http"
+	"strings"
 )

 func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode {
@ -54,7 +53,7 @@ func RelayAudioHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		}
 	}

-	modelRatio := billingratio.GetModelRatio(audioModel, channelType)
+	modelRatio := billingratio.GetModelRatio(audioModel)
 	groupRatio := billingratio.GetGroupRatio(group)
 	ratio := modelRatio * groupRatio
 	var quota int64
--- a/relay/controller/helper.go
+++ b/relay/controller/helper.go
@ -4,10 +4,6 @@ import (
 	"context"
 	"errors"
 	"fmt"
-	"math"
-	"net/http"
-	"strings"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
@ -20,6 +16,9 @@ import (
 	"github.com/songquanpeng/one-api/relay/meta"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
+	"math"
+	"net/http"
+	"strings"
 )

 func getAndValidateTextRequest(c *gin.Context, relayMode int) (*relaymodel.GeneralOpenAIRequest, error) {
@ -41,6 +40,78 @@ func getAndValidateTextRequest(c *gin.Context, relayMode int) (*relaymodel.Gener
 	return textRequest, nil
 }

+func getImageRequest(c *gin.Context, relayMode int) (*relaymodel.ImageRequest, error) {
+	imageRequest := &relaymodel.ImageRequest{}
+	err := common.UnmarshalBodyReusable(c, imageRequest)
+	if err != nil {
+		return nil, err
+	}
+	if imageRequest.N == 0 {
+		imageRequest.N = 1
+	}
+	if imageRequest.Size == "" {
+		imageRequest.Size = "1024x1024"
+	}
+	if imageRequest.Model == "" {
+		imageRequest.Model = "dall-e-2"
+	}
+	return imageRequest, nil
+}
+
+func isValidImageSize(model string, size string) bool {
+	if model == "cogview-3" {
+		return true
+	}
+	_, ok := billingratio.ImageSizeRatios[model][size]
+	return ok
+}
+
+func getImageSizeRatio(model string, size string) float64 {
+	ratio, ok := billingratio.ImageSizeRatios[model][size]
+	if !ok {
+		return 1
+	}
+	return ratio
+}
+
+func validateImageRequest(imageRequest *relaymodel.ImageRequest, meta *meta.Meta) *relaymodel.ErrorWithStatusCode {
+	// model validation
+	hasValidSize := isValidImageSize(imageRequest.Model, imageRequest.Size)
+	if !hasValidSize {
+		return openai.ErrorWrapper(errors.New("size not supported for this image model"), "size_not_supported", http.StatusBadRequest)
+	}
+	// check prompt length
+	if imageRequest.Prompt == "" {
+		return openai.ErrorWrapper(errors.New("prompt is required"), "prompt_missing", http.StatusBadRequest)
+	}
+	if len(imageRequest.Prompt) > billingratio.ImagePromptLengthLimitations[imageRequest.Model] {
+		return openai.ErrorWrapper(errors.New("prompt is too long"), "prompt_too_long", http.StatusBadRequest)
+	}
+	// Number of generated images validation
+	if !isWithinRange(imageRequest.Model, imageRequest.N) {
+		// channel not azure
+		if meta.ChannelType != channeltype.Azure {
+			return openai.ErrorWrapper(errors.New("invalid value of n"), "n_not_within_range", http.StatusBadRequest)
+		}
+	}
+	return nil
+}
+
+func getImageCostRatio(imageRequest *relaymodel.ImageRequest) (float64, error) {
+	if imageRequest == nil {
+		return 0, errors.New("imageRequest is nil")
+	}
+	imageCostRatio := getImageSizeRatio(imageRequest.Model, imageRequest.Size)
+	if imageRequest.Quality == "hd" && imageRequest.Model == "dall-e-3" {
+		if imageRequest.Size == "1024x1024" {
+			imageCostRatio *= 2
+		} else {
+			imageCostRatio *= 1.5
+		}
+	}
+	return imageCostRatio, nil
+}
+
 func getPromptTokens(textRequest *relaymodel.GeneralOpenAIRequest, relayMode int) int {
 	switch relayMode {
 	case relaymode.ChatCompletions:
@ -96,7 +167,7 @@ func postConsumeQuota(ctx context.Context, usage *relaymodel.Usage, meta *meta.M
 		return
 	}
 	var quota int64
-	completionRatio := billingratio.GetCompletionRatio(textRequest.Model, meta.ChannelType)
+	completionRatio := billingratio.GetCompletionRatio(textRequest.Model)
 	promptTokens := usage.PromptTokens
 	completionTokens := usage.CompletionTokens
 	quota = int64(math.Ceil((float64(promptTokens) + float64(completionTokens)*completionRatio) * ratio))
--- a/relay/controller/image.go
+++ b/relay/controller/image.go
@ -6,11 +6,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"io"
-	"net/http"
-
 	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/model"
@ -20,86 +16,17 @@ import (
 	"github.com/songquanpeng/one-api/relay/channeltype"
 	"github.com/songquanpeng/one-api/relay/meta"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
 )

-func getImageRequest(c *gin.Context, relayMode int) (*relaymodel.ImageRequest, error) {
-	imageRequest := &relaymodel.ImageRequest{}
-	err := common.UnmarshalBodyReusable(c, imageRequest)
-	if err != nil {
-		return nil, err
-	}
-	if imageRequest.N == 0 {
-		imageRequest.N = 1
-	}
-	if imageRequest.Size == "" {
-		imageRequest.Size = "1024x1024"
-	}
-	if imageRequest.Model == "" {
-		imageRequest.Model = "dall-e-2"
-	}
-	return imageRequest, nil
-}
-
-func isValidImageSize(model string, size string) bool {
-	if model == "cogview-3" || billingratio.ImageSizeRatios[model] == nil {
-		return true
-	}
-	_, ok := billingratio.ImageSizeRatios[model][size]
-	return ok
-}
-
-func isValidImagePromptLength(model string, promptLength int) bool {
-	maxPromptLength, ok := billingratio.ImagePromptLengthLimitations[model]
-	return !ok || promptLength <= maxPromptLength
-}
-
 func isWithinRange(element string, value int) bool {
-	amounts, ok := billingratio.ImageGenerationAmounts[element]
-	return !ok || (value >= amounts[0] && value <= amounts[1])
-}
-
-func getImageSizeRatio(model string, size string) float64 {
-	if ratio, ok := billingratio.ImageSizeRatios[model][size]; ok {
-		return ratio
+	if _, ok := billingratio.ImageGenerationAmounts[element]; !ok {
+		return false
 	}
-	return 1
-}
-
-func validateImageRequest(imageRequest *relaymodel.ImageRequest, meta *meta.Meta) *relaymodel.ErrorWithStatusCode {
-	// check prompt length
-	if imageRequest.Prompt == "" {
-		return openai.ErrorWrapper(errors.New("prompt is required"), "prompt_missing", http.StatusBadRequest)
-	}
-
-	// model validation
-	if !isValidImageSize(imageRequest.Model, imageRequest.Size) {
-		return openai.ErrorWrapper(errors.New("size not supported for this image model"), "size_not_supported", http.StatusBadRequest)
-	}
-
-	if !isValidImagePromptLength(imageRequest.Model, len(imageRequest.Prompt)) {
-		return openai.ErrorWrapper(errors.New("prompt is too long"), "prompt_too_long", http.StatusBadRequest)
-	}
-
-	// Number of generated images validation
-	if !isWithinRange(imageRequest.Model, imageRequest.N) {
-		return openai.ErrorWrapper(errors.New("invalid value of n"), "n_not_within_range", http.StatusBadRequest)
-	}
-	return nil
-}
-
-func getImageCostRatio(imageRequest *relaymodel.ImageRequest) (float64, error) {
-	if imageRequest == nil {
-		return 0, errors.New("imageRequest is nil")
-	}
-	imageCostRatio := getImageSizeRatio(imageRequest.Model, imageRequest.Size)
-	if imageRequest.Quality == "hd" && imageRequest.Model == "dall-e-3" {
-		if imageRequest.Size == "1024x1024" {
-			imageCostRatio *= 2
-		} else {
-			imageCostRatio *= 1.5
-		}
-	}
-	return imageCostRatio, nil
+	min := billingratio.ImageGenerationAmounts[element][0]
+	max := billingratio.ImageGenerationAmounts[element][1]
+	return value >= min && value <= max
 }

 func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode {
@ -167,7 +94,7 @@ func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatus
 		requestBody = bytes.NewBuffer(jsonStr)
 	}

-	modelRatio := billingratio.GetModelRatio(imageModel, meta.ChannelType)
+	modelRatio := billingratio.GetModelRatio(imageModel)
 	groupRatio := billingratio.GetGroupRatio(meta.Group)
 	ratio := modelRatio * groupRatio
 	userQuota, err := model.CacheGetUserQuota(ctx, meta.UserId)
--- a/relay/controller/proxy.go
+++ b/relay/controller/proxy.go
@ -1,41 +0,0 @@
-// Package controller is a package for handling the relay controller
-package controller
-
-import (
-	"fmt"
-	"net/http"
-
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/logger"
-	"github.com/songquanpeng/one-api/relay"
-	"github.com/songquanpeng/one-api/relay/adaptor/openai"
-	"github.com/songquanpeng/one-api/relay/meta"
-	relaymodel "github.com/songquanpeng/one-api/relay/model"
-)
-
-// RelayProxyHelper is a helper function to proxy the request to the upstream service
-func RelayProxyHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode {
-	ctx := c.Request.Context()
-	meta := meta.GetByContext(c)
-
-	adaptor := relay.GetAdaptor(meta.APIType)
-	if adaptor == nil {
-		return openai.ErrorWrapper(fmt.Errorf("invalid api type: %d", meta.APIType), "invalid_api_type", http.StatusBadRequest)
-	}
-	adaptor.Init(meta)
-
-	resp, err := adaptor.DoRequest(c, meta, c.Request.Body)
-	if err != nil {
-		logger.Errorf(ctx, "DoRequest failed: %s", err.Error())
-		return openai.ErrorWrapper(err, "do_request_failed", http.StatusInternalServerError)
-	}
-
-	// do response
-	_, respErr := adaptor.DoResponse(c, resp, meta)
-	if respErr != nil {
-		logger.Errorf(ctx, "respErr is not nil: %+v", respErr)
-		return respErr
-	}
-
-	return nil
-}
--- a/relay/controller/text.go
+++ b/relay/controller/text.go
@ -4,13 +4,9 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"io"
-	"net/http"
-
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/relay"
-	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
 	"github.com/songquanpeng/one-api/relay/apitype"
 	"github.com/songquanpeng/one-api/relay/billing"
@ -18,6 +14,8 @@ import (
 	"github.com/songquanpeng/one-api/relay/channeltype"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
 )

 func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
@ -32,11 +30,12 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 	meta.IsStream = textRequest.Stream

 	// map model name
+	var isModelMapped bool
 	meta.OriginModelName = textRequest.Model
-	textRequest.Model, _ = getMappedModelName(textRequest.Model, meta.ModelMapping)
+	textRequest.Model, isModelMapped = getMappedModelName(textRequest.Model, meta.ModelMapping)
 	meta.ActualModelName = textRequest.Model
 	// get model ratio & group ratio
-	modelRatio := billingratio.GetModelRatio(textRequest.Model, meta.ChannelType)
+	modelRatio := billingratio.GetModelRatio(textRequest.Model)
 	groupRatio := billingratio.GetGroupRatio(meta.Group)
 	ratio := modelRatio * groupRatio
 	// pre-consume quota
@ -55,9 +54,30 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 	adaptor.Init(meta)

 	// get request body
-	requestBody, err := getRequestBody(c, meta, textRequest, adaptor)
-	if err != nil {
-		return openai.ErrorWrapper(err, "convert_request_failed", http.StatusInternalServerError)
+	var requestBody io.Reader
+	if meta.APIType == apitype.OpenAI {
+		// no need to convert request for openai
+		shouldResetRequestBody := isModelMapped || meta.ChannelType == channeltype.Baichuan // frequency_penalty 0 is not acceptable for baichuan
+		if shouldResetRequestBody {
+			jsonStr, err := json.Marshal(textRequest)
+			if err != nil {
+				return openai.ErrorWrapper(err, "json_marshal_failed", http.StatusInternalServerError)
+			}
+			requestBody = bytes.NewBuffer(jsonStr)
+		} else {
+			requestBody = c.Request.Body
+		}
+	} else {
+		convertedRequest, err := adaptor.ConvertRequest(c, meta.Mode, textRequest)
+		if err != nil {
+			return openai.ErrorWrapper(err, "convert_request_failed", http.StatusInternalServerError)
+		}
+		jsonData, err := json.Marshal(convertedRequest)
+		if err != nil {
+			return openai.ErrorWrapper(err, "json_marshal_failed", http.StatusInternalServerError)
+		}
+		logger.Debugf(ctx, "converted request: \n%s", string(jsonData))
+		requestBody = bytes.NewBuffer(jsonData)
 	}

 	// do request
@ -82,26 +102,3 @@ func RelayTextHelper(c *gin.Context) *model.ErrorWithStatusCode {
 	go postConsumeQuota(ctx, usage, meta, textRequest, ratio, preConsumedQuota, modelRatio, groupRatio)
 	return nil
 }
-
-func getRequestBody(c *gin.Context, meta *meta.Meta, textRequest *model.GeneralOpenAIRequest, adaptor adaptor.Adaptor) (io.Reader, error) {
-	if meta.APIType == apitype.OpenAI && meta.OriginModelName == meta.ActualModelName && meta.ChannelType != channeltype.Baichuan {
-		// no need to convert request for openai
-		return c.Request.Body, nil
-	}
-
-	// get request body
-	var requestBody io.Reader
-	convertedRequest, err := adaptor.ConvertRequest(c, meta.Mode, textRequest)
-	if err != nil {
-		logger.Debugf(c.Request.Context(), "converted request failed: %s\n", err.Error())
-		return nil, err
-	}
-	jsonData, err := json.Marshal(convertedRequest)
-	if err != nil {
-		logger.Debugf(c.Request.Context(), "converted request json_marshal_failed: %s\n", err.Error())
-		return nil, err
-	}
-	logger.Debugf(c.Request.Context(), "converted request: \n%s", string(jsonData))
-	requestBody = bytes.NewBuffer(jsonData)
-	return requestBody, nil
-}
--- a/relay/meta/relay_meta.go
+++ b/relay/meta/relay_meta.go
@ -10,23 +10,20 @@ import (
 )

 type Meta struct {
-	Mode         int
-	ChannelType  int
-	ChannelId    int
-	TokenId      int
-	TokenName    string
-	UserId       int
-	Group        string
-	ModelMapping map[string]string
-	// BaseURL is the proxy url set in the channel config
-	BaseURL  string
-	APIKey   string
-	APIType  int
-	Config   model.ChannelConfig
-	IsStream bool
-	// OriginModelName is the model name from the raw user request
+	Mode            int
+	ChannelType     int
+	ChannelId       int
+	TokenId         int
+	TokenName       string
+	UserId          int
+	Group           string
+	ModelMapping    map[string]string
+	BaseURL         string
+	APIKey          string
+	APIType         int
+	Config          model.ChannelConfig
+	IsStream        bool
 	OriginModelName string
-	// ActualModelName is the model name after mapping
 	ActualModelName string
 	RequestURLPath  string
 	PromptTokens    int // only for DoResponse
--- a/relay/model/general.go
+++ b/relay/model/general.go
@ -1,15 +1,7 @@
 package model

 type ResponseFormat struct {
-	Type       string      `json:"type,omitempty"`
-	JsonSchema *JSONSchema `json:"json_schema,omitempty"`
-}
-
-type JSONSchema struct {
-	Description string                 `json:"description,omitempty"`
-	Name        string                 `json:"name"`
-	Schema      map[string]interface{} `json:"schema,omitempty"`
-	Strict      *bool                  `json:"strict,omitempty"`
+	Type string `json:"type,omitempty"`
 }

 type GeneralOpenAIRequest struct {
@ -21,7 +13,6 @@ type GeneralOpenAIRequest struct {
 	PresencePenalty  float64         `json:"presence_penalty,omitempty"`
 	ResponseFormat   *ResponseFormat `json:"response_format,omitempty"`
 	Seed             float64         `json:"seed,omitempty"`
-	Stop             any             `json:"stop,omitempty"`
 	Stream           bool            `json:"stream,omitempty"`
 	Temperature      float64         `json:"temperature,omitempty"`
 	TopP             float64         `json:"top_p,omitempty"`
@ -37,7 +28,6 @@ type GeneralOpenAIRequest struct {
 	Dimensions       int             `json:"dimensions,omitempty"`
 	Instruction      string          `json:"instruction,omitempty"`
 	Size             string          `json:"size,omitempty"`
-	NumCtx           int         	 `json:"num_ctx,omitempty"`
 }

 func (r GeneralOpenAIRequest) ParseInput() []string {
--- a/relay/model/message.go
+++ b/relay/model/message.go
@ -1,11 +1,10 @@
 package model

 type Message struct {
-	Role       string  `json:"role,omitempty"`
-	Content    any     `json:"content,omitempty"`
-	Name       *string `json:"name,omitempty"`
-	ToolCalls  []Tool  `json:"tool_calls,omitempty"`
-	ToolCallId string  `json:"tool_call_id,omitempty"`
+	Role      string  `json:"role,omitempty"`
+	Content   any     `json:"content,omitempty"`
+	Name      *string `json:"name,omitempty"`
+	ToolCalls []Tool  `json:"tool_calls,omitempty"`
 }

 func (m Message) IsStringContent() bool {
--- a/relay/model/tool.go
+++ b/relay/model/tool.go
@ -2,13 +2,13 @@ package model

 type Tool struct {
 	Id       string   `json:"id,omitempty"`
-	Type     string   `json:"type,omitempty"` // when splicing claude tools stream messages, it is empty
+	Type     string   `json:"type"`
 	Function Function `json:"function"`
 }

 type Function struct {
 	Description string `json:"description,omitempty"`
-	Name        string `json:"name,omitempty"`       // when splicing claude tools stream messages, it is empty
+	Name        string `json:"name"`
 	Parameters  any    `json:"parameters,omitempty"` // request
 	Arguments   any    `json:"arguments,omitempty"`  // response
 }
--- a/Show More
+++ b/Show More