feat: support multipart/form-data format request (#1690)

* "add parser multipart/form-data"

* chore: fix impl

* chore: update impl

---------

Co-authored-by: JustSong <songquanpeng@foxmail.com>

.env.example

@@ -0,0 +1,3 @@
+PORT=3000
+DEBUG=false
+HTTPS_PROXY=http://localhost:7890

.github/workflows/ci.yml

@@ -0,0 +1,47 @@
+name: CI
+
+# This setup assumes that you run the unit tests with code coverage in the same
+# workflow that will also print the coverage report as comment to the pull request. 
+# Therefore, you need to trigger this workflow when a pull request is (re)opened or
+# when new code is pushed to the branch of the pull request. In addition, you also
+# need to trigger this workflow when new code is pushed to the main branch because 
+# we need to upload the code coverage results as artifact for the main branch as
+# well since it will be the baseline code coverage.
+# 
+# We do not want to trigger the workflow for pushes to *any* branch because this
+# would trigger our jobs twice on pull requests (once from "push" event and once
+# from "pull_request->synchronize")
+on:
+  pull_request:
+    types: [opened, reopened, synchronize]
+  push:
+    branches:
+      - 'main'
+
+jobs:
+  unit_tests:
+    name: "Unit tests"
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Setup Go
+        uses: actions/setup-go@v4
+        with:
+          go-version: ^1.22
+
+      # When you execute your unit tests, make sure to use the "-coverprofile" flag to write a 
+      # coverage profile to a file. You will need the name of the file (e.g. "coverage.txt")
+      # in the next step as well as the next job.
+      - name: Test
+        run: go test -cover -coverprofile=coverage.txt ./...
+      - uses: codecov/codecov-action@v4
+        with:
+          token: ${{ secrets.CODECOV_TOKEN }}
+
+  commit_lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: wagoid/commitlint-github-action@v6

.github/workflows/docker-image-amd64.yml

@@ -1,61 +0,0 @@
-name: Publish Docker image (amd64)
-
-on:
-  push:
-    tags:
-      - '*'
-  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 }}

.github/workflows/docker-image-en.yml

@@ -1,9 +1,9 @@
-name: Publish Docker image (amd64, English)
+name: Publish Docker image (English)
 
 on:
   push:
     tags:
-      - '*'
+      - 'v*.*.*'
   workflow_dispatch:
     inputs:
       name:
@@ -34,6 +34,13 @@ 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:
@@ -51,6 +58,7 @@ 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 }}

.github/workflows/docker-image.yml

@@ -1,10 +1,9 @@
-name: Publish Docker image (arm64)
+name: Publish Docker image
 
 on:
   push:
     tags:
-      - '*'
-      - '!*-alpha*'
+      - 'v*.*.*'
   workflow_dispatch:
     inputs:
       name:

.github/workflows/linux-release.yml

@@ -5,7 +5,7 @@ permissions:
 on:
   push:
     tags:
-      - '*'
+      - 'v*.*.*'
       - '!*-alpha*'
   workflow_dispatch:
     inputs:

.github/workflows/macos-release.yml

@@ -5,7 +5,7 @@ permissions:
 on:
   push:
     tags:
-      - '*'
+      - 'v*.*.*'
       - '!*-alpha*'
   workflow_dispatch:
     inputs:

.github/workflows/windows-release.yml

@@ -5,7 +5,7 @@ permissions:
 on:
   push:
     tags:
-      - '*'
+      - 'v*.*.*'
       - '!*-alpha*'
   workflow_dispatch:
     inputs:

.gitignore

@@ -9,3 +9,4 @@ logs
 data
 /web/node_modules
 cmd.md
+.env

Dockerfile

@@ -1,4 +1,4 @@
-FROM node:16 as builder
+FROM --platform=$BUILDPLATFORM node:16 AS builder
 
 WORKDIR /web
 COPY ./VERSION .
@@ -16,7 +16,9 @@ WORKDIR /web/air
 RUN npm install
 RUN DISABLE_ESLINT_PLUGIN='true' REACT_APP_VERSION=$(cat VERSION) npm run build
 
-FROM golang AS builder2
+FROM golang:alpine AS builder2
+
+RUN apk add --no-cache g++
 
 ENV GO111MODULE=on \
     CGO_ENABLED=1 \
@@ -27,7 +29,7 @@ ADD go.mod go.sum ./
 RUN go mod download
 COPY . .
 COPY --from=builder /web/build ./web/build
-RUN go build -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)' -extldflags '-static'" -o one-api
+RUN go build -trimpath -ldflags "-s -w -X 'github.com/songquanpeng/one-api/common.Version=$(cat VERSION)' -extldflags '-static'" -o one-api
 
 FROM alpine
 

README.en.md

@@ -245,16 +245,41 @@ 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. `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. '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.
     + Example: `SYNC_FREQUENCY=60`
-7. `NODE_TYPE`: When set, specifies the node type. Valid values are `master` and `slave`. If not set, it defaults to `master`.
+8. `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`
-8. `CHANNEL_UPDATE_FREQUENCY`: When set, it periodically updates the channel balances, with the unit in minutes. If not set, no update will happen.
+9. `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`
-9. `CHANNEL_TEST_FREQUENCY`: When set, it periodically tests the channels, with the unit in minutes. If not set, no test will happen.
+10. `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`
-10. `POLLING_INTERVAL`: The time interval (in seconds) between requests when updating channel balances and testing channel availability. Default is no interval.
+11. `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`.
@@ -287,7 +312,9 @@ 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
+* [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.
 
 ## 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.

README.md

@@ -68,6 +68,7 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
    + [x] [Anthropic Claude 系列模型](https://anthropic.com) (支持 AWS Claude)
    + [x] [Google PaLM2/Gemini 系列模型](https://developers.generativeai.google)
    + [x] [Mistral 系列模型](https://mistral.ai/)
+   + [x] [字节跳动豆包大模型](https://console.volcengine.com/ark/region:ark+cn-beijing/model)
    + [x] [百度文心一言系列模型](https://cloud.baidu.com/doc/WENXINWORKSHOP/index.html)
    + [x] [阿里通义千问系列模型](https://help.aliyun.com/document_detail/2400395.html)
    + [x] [讯飞星火认知大模型](https://www.xfyun.cn/doc/spark/Web.html)
@@ -76,12 +77,19 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用
    + [x] [腾讯混元大模型](https://cloud.tencent.com/document/product/1729)
    + [x] [Moonshot AI](https://platform.moonshot.cn/)
    + [x] [百川大模型](https://platform.baichuan-ai.com)
-   + [ ] [字节云雀大模型](https://www.volcengine.com/product/ark) (WIP)
    + [x] [MINIMAX](https://api.minimax.chat/)
    + [x] [Groq](https://wow.groq.com/)
    + [x] [Ollama](https://github.com/ollama/ollama)
    + [x] [零一万物](https://platform.lingyiwanwu.com/)
    + [x] [阶跃星辰](https://platform.stepfun.com/)
+   + [x] [Coze](https://www.coze.com/)
+   + [x] [Cohere](https://cohere.com/)
+   + [x] [DeepSeek](https://www.deepseek.com/)
+   + [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。
@@ -244,9 +252,9 @@ 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。
+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。
 
-可安装 [Switcher 插件](https://github.com/RockChinQ/Switcher)在运行时切换所使用的模型。
+运行期间可以通过`!model`命令查看、切换可用模型。
 
 ### 部署到第三方平台
 <details>
@@ -334,6 +342,7 @@ graph LR
 不加的话将会使用负载均衡的方式使用多个渠道。
 
 ### 环境变量
+> One API 支持从 `.env` 文件中读取环境变量，请参照 `.env.example` 文件，使用时请将其重命名为 `.env`。
 1. `REDIS_CONN_STRING`：设置之后将使用 Redis 作为缓存使用。
    + 例子：`REDIS_CONN_STRING=redis://default:redispw@localhost:49153`
    + 如果数据库访问延迟很低，没有必要启用 Redis，启用后反而会出现数据滞后的问题。
@@ -363,29 +372,33 @@ graph LR
 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。 
-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：`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。
     + 例子：`POLLING_INTERVAL=5`
-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`。
+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`。
     + 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。
-14. `BATCH_UPDATE_INTERVAL=5`：批量更新聚合的时间间隔，单位为秒，默认为 `5`。
+13. `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`
-15. 请求频率限制：
+14. 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`。
-16. 编码器缓存设置：
+15. 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。
-17. `RELAY_TIMEOUT`：中继超时设置，单位为秒，默认不设置超时时间。
-18. `SQLITE_BUSY_TIMEOUT`：SQLite 锁等待超时设置，单位为毫秒，默认 `3000`。
-19. `GEMINI_SAFETY_SETTING`：Gemini 的安全设置，默认 `BLOCK_NONE`。
-20. `GEMINI_VERSION`：One API 所使用的 Gemini 版本，默认为 `v1`。
-21. `THEME`：系统的主题设置，默认为 `default`，具体可选值参考[此处](./web/README.md)。
-22. `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`。
-23. `METRIC_QUEUE_SIZE`：请求成功率统计队列大小，默认为 `10`。
-24. `METRIC_SUCCESS_RATE_THRESHOLD`：请求成功率阈值，默认为 `0.8`。
-25. `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。
+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。
 
 ### 命令行参数
 1. `--port <port_number>`: 指定服务器监听的端口号，默认为 `3000`。
@@ -438,6 +451,8 @@ 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
 

common/client/init.go

@@ -0,0 +1,60 @@
+package client
+
+import (
+	"fmt"
+	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/logger"
+	"net/http"
+	"net/url"
+	"time"
+)
+
+var HTTPClient *http.Client
+var ImpatientHTTPClient *http.Client
+var UserContentRequestHTTPClient *http.Client
+
+func Init() {
+	if config.UserContentRequestProxy != "" {
+		logger.SysLog(fmt.Sprintf("using %s as proxy to fetch user content", config.UserContentRequestProxy))
+		proxyURL, err := url.Parse(config.UserContentRequestProxy)
+		if err != nil {
+			logger.FatalLog(fmt.Sprintf("USER_CONTENT_REQUEST_PROXY set but invalid: %s", config.UserContentRequestProxy))
+		}
+		transport := &http.Transport{
+			Proxy: http.ProxyURL(proxyURL),
+		}
+		UserContentRequestHTTPClient = &http.Client{
+			Transport: transport,
+			Timeout:   time.Second * time.Duration(config.UserContentRequestTimeout),
+		}
+	} else {
+		UserContentRequestHTTPClient = &http.Client{}
+	}
+	var transport http.RoundTripper
+	if config.RelayProxy != "" {
+		logger.SysLog(fmt.Sprintf("using %s as api relay proxy", config.RelayProxy))
+		proxyURL, err := url.Parse(config.RelayProxy)
+		if err != nil {
+			logger.FatalLog(fmt.Sprintf("USER_CONTENT_REQUEST_PROXY set but invalid: %s", config.UserContentRequestProxy))
+		}
+		transport = &http.Transport{
+			Proxy: http.ProxyURL(proxyURL),
+		}
+	}
+
+	if config.RelayTimeout == 0 {
+		HTTPClient = &http.Client{
+			Transport: transport,
+		}
+	} else {
+		HTTPClient = &http.Client{
+			Timeout:   time.Duration(config.RelayTimeout) * time.Second,
+			Transport: transport,
+		}
+	}
+
+	ImpatientHTTPClient = &http.Client{
+		Timeout:   5 * time.Second,
+		Transport: transport,
+	}
+}

common/config/config.go

@@ -35,6 +35,7 @@ var PasswordLoginEnabled = true
 var PasswordRegisterEnabled = true
 var EmailVerificationEnabled = false
 var GitHubOAuthEnabled = false
+var OidcEnabled = false
 var WeChatAuthEnabled = false
 var TurnstileCheckEnabled = false
 var RegisterEnabled = true
@@ -70,6 +71,13 @@ var GitHubClientSecret = ""
 var LarkClientId = ""
 var LarkClientSecret = ""
 
+var OidcClientId = ""
+var OidcClientSecret = ""
+var OidcWellKnown = ""
+var OidcAuthorizationEndpoint = ""
+var OidcTokenEndpoint = ""
+var OidcUserinfoEndpoint = ""
+
 var WeChatServerAddress = ""
 var WeChatServerToken = ""
 var WeChatAccountQRCodeImageURL = ""
@@ -117,10 +125,10 @@ var ValidThemes = map[string]bool{
 // All duration's unit is seconds
 // Shouldn't larger then RateLimitKeyExpirationDuration
 var (
-	GlobalApiRateLimitNum            = env.Int("GLOBAL_API_RATE_LIMIT", 180)
+	GlobalApiRateLimitNum            = env.Int("GLOBAL_API_RATE_LIMIT", 240)
 	GlobalApiRateLimitDuration int64 = 3 * 60
 
-	GlobalWebRateLimitNum            = env.Int("GLOBAL_WEB_RATE_LIMIT", 60)
+	GlobalWebRateLimitNum            = env.Int("GLOBAL_WEB_RATE_LIMIT", 120)
 	GlobalWebRateLimitDuration int64 = 3 * 60
 
 	UploadRateLimitNum            = 10
@@ -143,4 +151,12 @@ 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)

common/config/key.go

@@ -1,12 +0,0 @@
-package config
-
-const (
-	KeyPrefix = "cfg_"
-
-	KeyAPIVersion = KeyPrefix + "api_version"
-	KeyLibraryID  = KeyPrefix + "library_id"
-	KeyPlugin     = KeyPrefix + "plugin"
-	KeySK         = KeyPrefix + "sk"
-	KeyAK         = KeyPrefix + "ak"
-	KeyRegion     = KeyPrefix + "region"
-)

common/ctxkey/key.go

@@ -1,7 +1,23 @@
 package ctxkey
 
-var (
+const (
+	Config            = "config"
+	Id                = "id"
+	Username          = "username"
+	Role              = "role"
+	Status            = "status"
+	Channel           = "channel"
+	ChannelId         = "channel_id"
+	SpecificChannelId = "specific_channel_id"
 	RequestModel      = "request_model"
 	ConvertedRequest  = "converted_request"
 	OriginalModel     = "original_model"
+	Group             = "group"
+	ModelMapping      = "model_mapping"
+	ChannelName       = "channel_name"
+	TokenId           = "token_id"
+	TokenName         = "token_name"
+	BaseURL           = "base_url"
+	AvailableModels   = "available_models"
+	KeyRequestBody    = "key_request_body"
 )

common/gin.go

@@ -4,14 +4,13 @@ import (
 	"bytes"
 	"encoding/json"
 	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"io"
 	"strings"
 )
 
-const KeyRequestBody = "key_request_body"
-
 func GetRequestBody(c *gin.Context) ([]byte, error) {
-	requestBody, _ := c.Get(KeyRequestBody)
+	requestBody, _ := c.Get(ctxkey.KeyRequestBody)
 	if requestBody != nil {
 		return requestBody.([]byte), nil
 	}
@@ -20,7 +19,7 @@ func GetRequestBody(c *gin.Context) ([]byte, error) {
 		return nil, err
 	}
 	_ = c.Request.Body.Close()
-	c.Set(KeyRequestBody, requestBody)
+	c.Set(ctxkey.KeyRequestBody, requestBody)
 	return requestBody.([]byte), nil
 }
 
@@ -32,15 +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 {
-		// skip for now
-		// TODO: someday non json request have variant model, we will need to implementation this
+		c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
+		err = c.ShouldBind(&v)
 	}
 	if err != nil {
 		return err
 	}
 	// Reset request body
-	c.Request.Body = io.NopCloser(bytes.NewBuffer(requestBody))
 	return nil
 }
 

common/helper/helper.go

@@ -2,6 +2,7 @@ package helper
 
 import (
 	"fmt"
+	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/random"
 	"html/template"
 	"log"
@@ -105,6 +106,11 @@ func GenRequestID() string {
 	return GetTimeString() + random.GetRandomNumberString(8)
 }
 
+func GetResponseID(c *gin.Context) string {
+	logID := c.GetString(RequestIdKey)
+	return fmt.Sprintf("chatcmpl-%s", logID)
+}
+
 func Max(a int, b int) int {
 	if a >= b {
 		return a

common/helper/key.go

@@ -0,0 +1,5 @@
+package helper
+
+const (
+	RequestIdKey = "X-Oneapi-Request-Id"
+)

common/image/image.go

@@ -3,6 +3,7 @@ package image
 import (
 	"bytes"
 	"encoding/base64"
+	"github.com/songquanpeng/one-api/common/client"
 	"image"
 	_ "image/gif"
 	_ "image/jpeg"
@@ -19,7 +20,7 @@ import (
 var dataURLPattern = regexp.MustCompile(`data:image/([^;]+);base64,(.*)`)
 
 func IsImageUrl(url string) (bool, error) {
-	resp, err := http.Head(url)
+	resp, err := client.UserContentRequestHTTPClient.Head(url)
 	if err != nil {
 		return false, err
 	}
@@ -34,7 +35,7 @@ func GetImageSizeFromUrl(url string) (width int, height int, err error) {
 	if !isImage {
 		return
 	}
-	resp, err := http.Get(url)
+	resp, err := client.UserContentRequestHTTPClient.Get(url)
 	if err != nil {
 		return
 	}

common/image/image_test.go

@@ -2,6 +2,7 @@ package image_test
 
 import (
 	"encoding/base64"
+	"github.com/songquanpeng/one-api/common/client"
 	"image"
 	_ "image/gif"
 	_ "image/jpeg"
@@ -44,6 +45,11 @@ var (
 	}
 )
 
+func TestMain(m *testing.M) {
+	client.Init()
+	m.Run()
+}
+
 func TestDecode(t *testing.T) {
 	// Bytes read: varies sometimes
 	// jpeg: 1063892

common/init.go

@@ -24,7 +24,7 @@ func printHelp() {
 	fmt.Println("Usage: one-api [--port <port>] [--log-dir <log directory>] [--version] [--help]")
 }
 
-func init() {
+func Init() {
 	flag.Parse()
 
 	if *PrintVersion {

common/logger/constants.go

@@ -1,7 +1,3 @@
 package logger
 
-const (
-	RequestIdKey = "X-Oneapi-Request-Id"
-)
-
 var LogDir string

common/logger/logger.go

@@ -3,15 +3,16 @@ package logger
 import (
 	"context"
 	"fmt"
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/config"
-	"github.com/songquanpeng/one-api/common/helper"
 	"io"
 	"log"
 	"os"
 	"path/filepath"
 	"sync"
 	"time"
+
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/helper"
 )
 
 const (
@@ -21,21 +22,17 @@ const (
 	loggerError = "ERR"
 )
 
-var setupLogLock sync.Mutex
-var setupLogWorking bool
+var setupLogOnce sync.Once
 
 func SetupLogger() {
+	setupLogOnce.Do(func() {
 		if LogDir != "" {
-		ok := setupLogLock.TryLock()
-		if !ok {
-			log.Println("setup log is already working")
-			return
+			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")))
 			}
-		defer func() {
-			setupLogLock.Unlock()
-			setupLogWorking = false
-		}()
-		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")
@@ -43,6 +40,7 @@ func SetupLogger() {
 			gin.DefaultWriter = io.MultiWriter(os.Stdout, fd)
 			gin.DefaultErrorWriter = io.MultiWriter(os.Stderr, fd)
 		}
+	})
 }
 
 func SysLog(s string) {
@@ -50,11 +48,19 @@ func SysLog(s string) {
 	_, _ = fmt.Fprintf(gin.DefaultWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s)
 }
 
+func SysLogf(format string, a ...any) {
+	SysLog(fmt.Sprintf(format, a...))
+}
+
 func SysError(s string) {
 	t := time.Now()
 	_, _ = fmt.Fprintf(gin.DefaultErrorWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s)
 }
 
+func SysErrorf(format string, a ...any) {
+	SysError(fmt.Sprintf(format, a...))
+}
+
 func Debug(ctx context.Context, msg string) {
 	if config.DebugEnabled {
 		logHelper(ctx, loggerDEBUG, msg)
@@ -94,18 +100,13 @@ func logHelper(ctx context.Context, level string, msg string) {
 	if level == loggerINFO {
 		writer = gin.DefaultWriter
 	}
-	id := ctx.Value(RequestIdKey)
+	id := ctx.Value(helper.RequestIdKey)
 	if id == nil {
 		id = helper.GenRequestID()
 	}
 	now := time.Now()
 	_, _ = fmt.Fprintf(writer, "[%s] %v | %s | %s \n", level, now.Format("2006/01/02 - 15:04:05"), id, msg)
-	if !setupLogWorking {
-		setupLogWorking = true
-		go func() {
 	SetupLogger()
-		}()
-	}
 }
 
 func FatalLog(v ...any) {

common/message/email.go

@@ -6,11 +6,16 @@ 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")
@@ -41,16 +46,24 @@ 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)
+			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 err != nil {
 			return err
 		}
@@ -59,9 +72,11 @@ 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.Mail(config.SMTPFrom); err != nil {
 			return err
 		}

common/render/render.go

@@ -0,0 +1,29 @@
+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]")
+}

controller/auth/oidc.go

@@ -0,0 +1,225 @@
+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
+}

controller/auth/wechat.go

@@ -6,6 +6,7 @@ import (
 	"fmt"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/controller"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
@@ -136,7 +137,7 @@ func WeChatBind(c *gin.Context) {
 		})
 		return
 	}
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	user := model.User{
 		Id: id,
 	}

controller/billing.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/model"
 	relaymodel "github.com/songquanpeng/one-api/relay/model"
 )
@@ -14,13 +15,15 @@ func GetSubscription(c *gin.Context) {
 	var token *model.Token
 	var expiredTime int64
 	if config.DisplayTokenStatEnabled {
-		tokenId := c.GetInt("token_id")
+		tokenId := c.GetInt(ctxkey.TokenId)
 		token, err = model.GetTokenById(tokenId)
+		if err == nil {
 			expiredTime = token.ExpiredTime
 			remainQuota = token.RemainQuota
 			usedQuota = token.UsedQuota
+		}
 	} else {
-		userId := c.GetInt("id")
+		userId := c.GetInt(ctxkey.Id)
 		remainQuota, err = model.GetUserQuota(userId)
 		if err != nil {
 			usedQuota, err = model.GetUserUsedQuota(userId)
@@ -64,11 +67,11 @@ func GetUsage(c *gin.Context) {
 	var err error
 	var token *model.Token
 	if config.DisplayTokenStatEnabled {
-		tokenId := c.GetInt("token_id")
+		tokenId := c.GetInt(ctxkey.TokenId)
 		token, err = model.GetTokenById(tokenId)
 		quota = token.UsedQuota
 	} else {
-		userId := c.GetInt("id")
+		userId := c.GetInt(ctxkey.Id)
 		quota, err = model.GetUserUsedQuota(userId)
 	}
 	if err != nil {

controller/channel-billing.go

@@ -4,12 +4,12 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"github.com/songquanpeng/one-api/common/client"
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/model"
 	"github.com/songquanpeng/one-api/monitor"
 	"github.com/songquanpeng/one-api/relay/channeltype"
-	"github.com/songquanpeng/one-api/relay/client"
 	"io"
 	"net/http"
 	"strconv"
@@ -81,6 +81,26 @@ 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{}
@@ -203,6 +223,28 @@ 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() == "" {
@@ -227,6 +269,8 @@ 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")
 	}

controller/channel-test.go

@@ -5,7 +5,18 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"net/url"
+	"strconv"
+	"strings"
+	"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"
 	"github.com/songquanpeng/one-api/common/message"
 	"github.com/songquanpeng/one-api/middleware"
@@ -17,23 +28,15 @@ 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"
-	"net/http/httptest"
-	"net/url"
-	"strconv"
-	"strings"
-	"sync"
-	"time"
-
-	"github.com/gin-gonic/gin"
 )
 
-func buildTestRequest() *relaymodel.GeneralOpenAIRequest {
+func buildTestRequest(model string) *relaymodel.GeneralOpenAIRequest {
+	if model == "" {
+		model = "gpt-3.5-turbo"
+	}
 	testRequest := &relaymodel.GeneralOpenAIRequest{
 		MaxTokens: 2,
-		Stream:    false,
-		Model:     "gpt-3.5-turbo",
+		Model:     model,
 	}
 	testMessage := relaymodel.Message{
 		Role:    "user",
@@ -43,7 +46,7 @@ func buildTestRequest() *relaymodel.GeneralOpenAIRequest {
 	return testRequest
 }
 
-func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error) {
+func testChannel(channel *model.Channel, request *relaymodel.GeneralOpenAIRequest) (err error, openaiErr *relaymodel.Error) {
 	w := httptest.NewRecorder()
 	c, _ := gin.CreateTestContext(w)
 	c.Request = &http.Request{
@@ -54,8 +57,10 @@ func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error
 	}
 	c.Request.Header.Set("Authorization", "Bearer "+channel.Key)
 	c.Request.Header.Set("Content-Type", "application/json")
-	c.Set("channel", channel.Type)
-	c.Set("base_url", channel.GetBaseURL())
+	c.Set(ctxkey.Channel, channel.Type)
+	c.Set(ctxkey.BaseURL, channel.GetBaseURL())
+	cfg, _ := channel.LoadConfig()
+	c.Set(ctxkey.Config, cfg)
 	middleware.SetupContextForSelectedChannel(c, channel, "")
 	meta := meta.GetByContext(c)
 	apiType := channeltype.ToAPIType(channel.Type)
@@ -64,16 +69,19 @@ func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error
 		return fmt.Errorf("invalid api type: %d, adaptor is nil", apiType), nil
 	}
 	adaptor.Init(meta)
-	modelName := adaptor.GetModelList()[0]
-	if !strings.Contains(channel.Models, modelName) {
+	modelName := request.Model
+	modelMap := channel.GetModelMapping()
+	if modelName == "" || !strings.Contains(channel.Models, modelName) {
 		modelNames := strings.Split(channel.Models, ",")
 		if len(modelNames) > 0 {
 			modelName = modelNames[0]
 		}
+		if modelMap != nil && modelMap[modelName] != "" {
+			modelName = modelMap[modelName]
 		}
-	request := buildTestRequest()
+	}
+	meta.OriginModelName, meta.ActualModelName = request.Model, modelName
 	request.Model = modelName
-	meta.OriginModelName, meta.ActualModelName = modelName, modelName
 	convertedRequest, err := adaptor.ConvertRequest(c, relaymode.ChatCompletions, request)
 	if err != nil {
 		return err, nil
@@ -82,6 +90,7 @@ func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error
 	if err != nil {
 		return err, nil
 	}
+	logger.SysLog(string(jsonData))
 	requestBody := bytes.NewBuffer(jsonData)
 	c.Request.Body = io.NopCloser(requestBody)
 	resp, err := adaptor.DoRequest(c, meta, requestBody)
@@ -126,10 +135,15 @@ func TestChannel(c *gin.Context) {
 		})
 		return
 	}
+	model := c.Query("model")
+	testRequest := buildTestRequest(model)
 	tik := time.Now()
-	err, _ = testChannel(channel)
+	err, _ = testChannel(channel, testRequest)
 	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 {
@@ -137,6 +151,7 @@ func TestChannel(c *gin.Context) {
 			"success": false,
 			"message": err.Error(),
 			"time":    consumedTime,
+			"model":   model,
 		})
 		return
 	}
@@ -144,6 +159,7 @@ func TestChannel(c *gin.Context) {
 		"success": true,
 		"message": "",
 		"time":    consumedTime,
+		"model":   model,
 	})
 	return
 }
@@ -174,11 +190,12 @@ func testChannels(notify bool, scope string) error {
 		for _, channel := range channels {
 			isChannelEnabled := channel.Status == model.ChannelStatusEnabled
 			tik := time.Now()
-			err, openaiErr := testChannel(channel)
+			testRequest := buildTestRequest("")
+			err, openaiErr := testChannel(channel, testRequest)
 			tok := time.Now()
 			milliseconds := tok.Sub(tik).Milliseconds()
 			if isChannelEnabled && milliseconds > disableThreshold {
-				err = errors.New(fmt.Sprintf("响应时间 %.2fs 超过阈值 %.2fs", float64(milliseconds)/1000.0, float64(disableThreshold)/1000.0))
+				err = fmt.Errorf("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 {

controller/log.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
 	"strconv"
@@ -41,7 +42,7 @@ func GetUserLogs(c *gin.Context) {
 	if p < 0 {
 		p = 0
 	}
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	logType, _ := strconv.Atoi(c.Query("type"))
 	startTimestamp, _ := strconv.ParseInt(c.Query("start_timestamp"), 10, 64)
 	endTimestamp, _ := strconv.ParseInt(c.Query("end_timestamp"), 10, 64)
@@ -83,7 +84,7 @@ func SearchAllLogs(c *gin.Context) {
 
 func SearchUserLogs(c *gin.Context) {
 	keyword := c.Query("keyword")
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	logs, err := model.SearchUserLogs(userId, keyword)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -122,7 +123,7 @@ func GetLogsStat(c *gin.Context) {
 }
 
 func GetLogsSelfStat(c *gin.Context) {
-	username := c.GetString("username")
+	username := c.GetString(ctxkey.Username)
 	logType, _ := strconv.Atoi(c.Query("type"))
 	startTimestamp, _ := strconv.ParseInt(c.Query("start_timestamp"), 10, 64)
 	endTimestamp, _ := strconv.ParseInt(c.Query("end_timestamp"), 10, 64)

controller/misc.go

@@ -36,6 +36,12 @@ func GetStatus(c *gin.Context) {
 			"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,
 		},
 	})
 	return

controller/model.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"fmt"
 	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/model"
 	relay "github.com/songquanpeng/one-api/relay"
 	"github.com/songquanpeng/one-api/relay/adaptor/openai"
@@ -131,10 +132,10 @@ func ListAllModels(c *gin.Context) {
 func ListModels(c *gin.Context) {
 	ctx := c.Request.Context()
 	var availableModels []string
-	if c.GetString("available_models") != "" {
-		availableModels = strings.Split(c.GetString("available_models"), ",")
+	if c.GetString(ctxkey.AvailableModels) != "" {
+		availableModels = strings.Split(c.GetString(ctxkey.AvailableModels), ",")
 	} else {
-		userId := c.GetInt("id")
+		userId := c.GetInt(ctxkey.Id)
 		userGroup, _ := model.CacheGetUserGroup(userId)
 		availableModels, _ = model.CacheGetGroupModels(ctx, userGroup)
 	}
@@ -186,7 +187,7 @@ func RetrieveModel(c *gin.Context) {
 
 func GetUserAvailableModels(c *gin.Context) {
 	ctx := c.Request.Context()
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	userGroup, err := model.CacheGetUserGroup(id)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{

controller/redemption.go

@@ -3,6 +3,7 @@ package controller
 import (
 	"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/helper"
 	"github.com/songquanpeng/one-api/common/random"
 	"github.com/songquanpeng/one-api/model"
@@ -109,7 +110,7 @@ func AddRedemption(c *gin.Context) {
 	for i := 0; i < redemption.Count; i++ {
 		key := random.GetUUID()
 		cleanRedemption := model.Redemption{
-			UserId:      c.GetInt("id"),
+			UserId:      c.GetInt(ctxkey.Id),
 			Name:        redemption.Name,
 			Key:         key,
 			CreatedTime: helper.GetTimestamp(),

controller/relay.go

@@ -4,6 +4,9 @@ import (
 	"bytes"
 	"context"
 	"fmt"
+	"io"
+	"net/http"
+
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
@@ -16,8 +19,6 @@ import (
 	"github.com/songquanpeng/one-api/relay/controller"
 	"github.com/songquanpeng/one-api/relay/model"
 	"github.com/songquanpeng/one-api/relay/relaymode"
-	"io"
-	"net/http"
 )
 
 // https://platform.openai.com/docs/api-reference/chat
@@ -33,6 +34,8 @@ 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)
 	}
@@ -46,18 +49,19 @@ func Relay(c *gin.Context) {
 		requestBody, _ := common.GetRequestBody(c)
 		logger.Debugf(ctx, "request body: %s", string(requestBody))
 	}
-	channelId := c.GetInt("channel_id")
+	channelId := c.GetInt(ctxkey.ChannelId)
+	userId := c.GetInt(ctxkey.Id)
 	bizErr := relayHelper(c, relayMode)
 	if bizErr == nil {
 		monitor.Emit(channelId, true)
 		return
 	}
 	lastFailedChannelId := channelId
-	channelName := c.GetString("channel_name")
-	group := c.GetString("group")
+	channelName := c.GetString(ctxkey.ChannelName)
+	group := c.GetString(ctxkey.Group)
 	originalModel := c.GetString(ctxkey.OriginalModel)
-	go processChannelRelayError(ctx, channelId, channelName, bizErr)
-	requestId := c.GetString(logger.RequestIdKey)
+	go processChannelRelayError(ctx, userId, channelId, channelName, bizErr)
+	requestId := c.GetString(helper.RequestIdKey)
 	retryTimes := config.RetryTimes
 	if !shouldRetry(c, bizErr.StatusCode) {
 		logger.Errorf(ctx, "relay error happen, status code is %d, won't retry in this case", bizErr.StatusCode)
@@ -80,15 +84,18 @@ func Relay(c *gin.Context) {
 		if bizErr == nil {
 			return
 		}
-		channelId := c.GetInt("channel_id")
+		channelId := c.GetInt(ctxkey.ChannelId)
 		lastFailedChannelId = channelId
-		channelName := c.GetString("channel_name")
-		go processChannelRelayError(ctx, channelId, channelName, bizErr)
+		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,
@@ -97,7 +104,7 @@ func Relay(c *gin.Context) {
 }
 
 func shouldRetry(c *gin.Context, statusCode int) bool {
-	if _, ok := c.Get("specific_channel_id"); ok {
+	if _, ok := c.Get(ctxkey.SpecificChannelId); ok {
 		return false
 	}
 	if statusCode == http.StatusTooManyRequests {
@@ -115,8 +122,8 @@ func shouldRetry(c *gin.Context, statusCode int) bool {
 	return true
 }
 
-func processChannelRelayError(ctx context.Context, channelId int, channelName string, err *model.ErrorWithStatusCode) {
-	logger.Errorf(ctx, "relay error (channel #%d): %s", channelId, err.Message)
+func processChannelRelayError(ctx context.Context, userId int, channelId int, channelName string, err *model.ErrorWithStatusCode) {
+	logger.Errorf(ctx, "relay error (channel id %d, user id: %d): %s", channelId, userId, err.Message)
 	// https://platform.openai.com/docs/guides/error-codes/api-errors
 	if monitor.ShouldDisableChannel(&err.Error, err.StatusCode) {
 		monitor.DisableChannel(channelId, channelName, err.Message)

controller/token.go

@@ -4,6 +4,7 @@ import (
 	"fmt"
 	"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/helper"
 	"github.com/songquanpeng/one-api/common/network"
 	"github.com/songquanpeng/one-api/common/random"
@@ -13,7 +14,7 @@ import (
 )
 
 func GetAllTokens(c *gin.Context) {
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	p, _ := strconv.Atoi(c.Query("p"))
 	if p < 0 {
 		p = 0
@@ -38,7 +39,7 @@ func GetAllTokens(c *gin.Context) {
 }
 
 func SearchTokens(c *gin.Context) {
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	keyword := c.Query("keyword")
 	tokens, err := model.SearchUserTokens(userId, keyword)
 	if err != nil {
@@ -58,7 +59,7 @@ func SearchTokens(c *gin.Context) {
 
 func GetToken(c *gin.Context) {
 	id, err := strconv.Atoi(c.Param("id"))
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -83,8 +84,8 @@ func GetToken(c *gin.Context) {
 }
 
 func GetTokenStatus(c *gin.Context) {
-	tokenId := c.GetInt("token_id")
-	userId := c.GetInt("id")
+	tokenId := c.GetInt(ctxkey.TokenId)
+	userId := c.GetInt(ctxkey.Id)
 	token, err := model.GetTokenByIds(tokenId, userId)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -139,7 +140,7 @@ func AddToken(c *gin.Context) {
 	}
 
 	cleanToken := model.Token{
-		UserId:         c.GetInt("id"),
+		UserId:         c.GetInt(ctxkey.Id),
 		Name:           token.Name,
 		Key:            random.GenerateKey(),
 		CreatedTime:    helper.GetTimestamp(),
@@ -168,7 +169,7 @@ func AddToken(c *gin.Context) {
 
 func DeleteToken(c *gin.Context) {
 	id, _ := strconv.Atoi(c.Param("id"))
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	err := model.DeleteTokenById(id, userId)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -185,7 +186,7 @@ func DeleteToken(c *gin.Context) {
 }
 
 func UpdateToken(c *gin.Context) {
-	userId := c.GetInt("id")
+	userId := c.GetInt(ctxkey.Id)
 	statusOnly := c.Query("status_only")
 	token := model.Token{}
 	err := c.ShouldBindJSON(&token)

controller/user.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/common/random"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
@@ -172,6 +173,7 @@ func Register(c *gin.Context) {
 		})
 		return
 	}
+
 	c.JSON(http.StatusOK, gin.H{
 		"success": true,
 		"message": "",
@@ -238,7 +240,7 @@ func GetUser(c *gin.Context) {
 		})
 		return
 	}
-	myRole := c.GetInt("role")
+	myRole := c.GetInt(ctxkey.Role)
 	if myRole <= user.Role && myRole != model.RoleRootUser {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -255,7 +257,7 @@ func GetUser(c *gin.Context) {
 }
 
 func GetUserDashboard(c *gin.Context) {
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	now := time.Now()
 	startOfDay := now.Truncate(24*time.Hour).AddDate(0, 0, -6).Unix()
 	endOfDay := now.Truncate(24 * time.Hour).Add(24*time.Hour - time.Second).Unix()
@@ -278,7 +280,7 @@ func GetUserDashboard(c *gin.Context) {
 }
 
 func GenerateAccessToken(c *gin.Context) {
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	user, err := model.GetUserById(id, true)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -314,7 +316,7 @@ func GenerateAccessToken(c *gin.Context) {
 }
 
 func GetAffCode(c *gin.Context) {
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	user, err := model.GetUserById(id, true)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -342,7 +344,7 @@ func GetAffCode(c *gin.Context) {
 }
 
 func GetSelf(c *gin.Context) {
-	id := c.GetInt("id")
+	id := c.GetInt(ctxkey.Id)
 	user, err := model.GetUserById(id, false)
 	if err != nil {
 		c.JSON(http.StatusOK, gin.H{
@@ -387,7 +389,7 @@ func UpdateUser(c *gin.Context) {
 		})
 		return
 	}
-	myRole := c.GetInt("role")
+	myRole := c.GetInt(ctxkey.Role)
 	if myRole <= originUser.Role && myRole != model.RoleRootUser {
 		c.JSON(http.StatusOK, gin.H{
 			"success": false,
@@ -445,7 +447,7 @@ func UpdateSelf(c *gin.Context) {
 	}
 
 	cleanUser := model.User{
-		Id:          c.GetInt("id"),
+		Id:          c.GetInt(ctxkey.Id),
 		Username:    user.Username,
 		Password:    user.Password,
 		DisplayName: user.DisplayName,

go.mod

@@ -4,53 +4,68 @@ module github.com/songquanpeng/one-api
 go 1.20
 
 require (
-	github.com/aws/aws-sdk-go-v2 v1.26.1
-	github.com/aws/aws-sdk-go-v2/credentials v1.17.11
-	github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.7.4
-	github.com/gin-contrib/cors v1.7.1
-	github.com/gin-contrib/gzip v1.0.0
-	github.com/gin-contrib/sessions v1.0.0
-	github.com/gin-contrib/static v1.1.1
-	github.com/gin-gonic/gin v1.9.1
-	github.com/go-playground/validator/v10 v10.19.0
+	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
+	github.com/gin-contrib/cors v1.7.2
+	github.com/gin-contrib/gzip v1.0.1
+	github.com/gin-contrib/sessions v1.0.1
+	github.com/gin-contrib/static v1.1.2
+	github.com/gin-gonic/gin v1.10.0
+	github.com/go-playground/validator/v10 v10.20.0
 	github.com/go-redis/redis/v8 v8.11.5
 	github.com/golang-jwt/jwt v3.2.2+incompatible
 	github.com/google/uuid v1.6.0
 	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 v0.1.6
+	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 v0.22.0
-	golang.org/x/image v0.15.0
+	golang.org/x/crypto v0.24.0
+	golang.org/x/image v0.18.0
+	google.golang.org/api v0.187.0
 	gorm.io/driver/mysql v1.5.6
 	gorm.io/driver/postgres v1.5.7
 	gorm.io/driver/sqlite v1.5.5
-	gorm.io/gorm v1.25.9
+	gorm.io/gorm v1.25.10
 )
 
 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 v1.3.5 // indirect
-	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.5 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7 // indirect
+	github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.7 // indirect
 	github.com/aws/smithy-go v1.20.2 // indirect
-	github.com/bytedance/sonic v1.11.5 // indirect
+	github.com/bytedance/sonic v1.11.6 // indirect
 	github.com/bytedance/sonic/loader v0.1.1 // indirect
 	github.com/cespare/xxhash/v2 v2.3.0 // indirect
-	github.com/cloudwego/base64x v0.1.3 // indirect
+	github.com/cloudwego/base64x v0.1.4 // indirect
 	github.com/cloudwego/iasm v0.2.0 // indirect
 	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 v0.10.2 // 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
@@ -67,19 +82,30 @@ 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 v2.0.3+incompatible // indirect
+	github.com/mattn/go-sqlite3 v1.14.22 // 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 v2.2.1 // indirect
+	github.com/pelletier/go-toml/v2 v2.2.2 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
 	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
-	golang.org/x/arch v0.7.0 // indirect
-	golang.org/x/net v0.24.0 // 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/sync v0.7.0 // indirect
-	golang.org/x/sys v0.19.0 // indirect
-	golang.org/x/text v0.14.0 // indirect
-	google.golang.org/protobuf v1.33.0 // 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
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 )

go.sum

@@ -1,30 +1,42 @@
+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/aws/aws-sdk-go-v2 v1.26.1 h1:5554eUqIYVWpU0YmeeYZ0wU64H2VLBs8TlhRB2L+EkA=
-github.com/aws/aws-sdk-go-v2 v1.26.1/go.mod h1:ffIFB97e2yNsv4aTSGkqtHnppsIJzw7G7BReUZ3jCXM=
+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=
 github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2/go.mod h1:lPprDr1e6cJdyYeGXnRaJoP4Md+cDBvi2eOj00BlGmg=
-github.com/aws/aws-sdk-go-v2/credentials v1.17.11 h1:YuIB1dJNf1Re822rriUOTxopaHHvIq0l/pX3fwO+Tzs=
-github.com/aws/aws-sdk-go-v2/credentials v1.17.11/go.mod h1:AQtFPsDH9bI2O+71anW6EKL+NcD7LG3dpKGMV4SShgo=
-github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.5 h1:aw39xVGeRWlWx9EzGVnhOR4yOjQDHPQ6o6NmBlscyQg=
-github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.5/go.mod h1:FSaRudD0dXiMPK2UjknVwwTYyZMRsHv3TtkabsZih5I=
-github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.5 h1:PG1F3OD1szkuQPzDw3CIQsRIrtTlUC3lP84taWzHlq0=
-github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.5/go.mod h1:jU1li6RFryMz+so64PpKtudI+QzbKoIEivqdf6LNpOc=
-github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.7.4 h1:JgHnonzbnA3pbqj76wYsSZIZZQYBxkmMEjvL6GHy8XU=
-github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.7.4/go.mod h1:nZspkhg+9p8iApLFoyAqfyuMP0F38acy2Hm3r5r95Cg=
+github.com/aws/aws-sdk-go-v2/credentials v1.17.15 h1:YDexlvDRCA8ems2T5IP1xkMtOZ1uLJOCJdTr0igs5zo=
+github.com/aws/aws-sdk-go-v2/credentials v1.17.15/go.mod h1:vxHggqW6hFNaeNC0WyXS3VdyjcV0a4KMUY4dKJ96buU=
+github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7 h1:lf/8VTF2cM+N4SLzaYJERKEWAXq8MOMpZfU6wEPWsPk=
+github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7/go.mod h1:4SjkU7QiqK2M9oozyMzfZ/23LmUY+h3oFqhdeP5OMiI=
+github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.7 h1:4OYVp0705xu8yjdyoWix0r9wPIRXnIzzOoUpQVHIJ/g=
+github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.7/go.mod h1:vd7ESTEvI76T2Na050gODNmNU7+OyKrIKroYTu4ABiI=
+github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.8.3 h1:Fihjyd6DeNjcawBEGLH9dkIEUi6AdhucDKPE9nJ4QiY=
+github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.8.3/go.mod h1:opvUj3ismqSCxYc+m4WIjPL0ewZGtvp0ess7cKvBPOQ=
 github.com/aws/smithy-go v1.20.2 h1:tbp628ireGtzcHDDmLT/6ADHidqnwgF57XOXZe6tp4Q=
 github.com/aws/smithy-go v1.20.2/go.mod h1:krry+ya/rV9RDcV/Q16kpu6ypI4K2czasz0NC3qS14E=
-github.com/bytedance/sonic v1.11.5 h1:G00FYjjqll5iQ1PYXynbg/hyzqBqavH8Mo9/oTopd9k=
-github.com/bytedance/sonic v1.11.5/go.mod h1:X2PC2giUdj/Cv2lliWFLk6c/DUQok5rViJSemeB0wDw=
-github.com/bytedance/sonic/loader v0.1.0/go.mod h1:UmRT+IRTGKz/DAkzcEGzyVqQFJ7H9BqwBO3pm9H/+HY=
+github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc0=
+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/cloudwego/base64x v0.1.3 h1:b5J/l8xolB7dyDTTmhJP2oTs5LdrjyrUFuNxdfq5hAg=
-github.com/cloudwego/base64x v0.1.3/go.mod h1:1+1K5BUHIQzyapgpF7LwvOGAEDicKtt1umPV+aN8pi8=
+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=
@@ -33,43 +45,83 @@ 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=
 github.com/gabriel-vasile/mimetype v1.4.3/go.mod h1:d8uq/6HKRL6CGdk+aubisF/M5GcPfT7nKyLpA0lbSSk=
-github.com/gin-contrib/cors v1.7.1 h1:s9SIppU/rk8enVvkzwiC2VK3UZ/0NNGsWfUKvV55rqs=
-github.com/gin-contrib/cors v1.7.1/go.mod h1:n/Zj7B4xyrgk/cX1WCX2dkzFfaNm/xJb6oIUk7WTtps=
-github.com/gin-contrib/gzip v1.0.0 h1:UKN586Po/92IDX6ie5CWLgMI81obiIp5nSP85T3wlTk=
-github.com/gin-contrib/gzip v1.0.0/go.mod h1:CtG7tQrPB3vIBo6Gat9FVUsis+1emjvQqd66ME5TdnE=
-github.com/gin-contrib/sessions v1.0.0 h1:r5GLta4Oy5xo9rAwMHx8B4wLpeRGHMdz9NafzJAdP8Y=
-github.com/gin-contrib/sessions v1.0.0/go.mod h1:DN0f4bvpqMQElDdi+gNGScrP2QEI04IErRyMFyorUOI=
+github.com/gin-contrib/cors v1.7.2 h1:oLDHxdg8W/XDoN/8zamqk/Drgt4oVZDvaV0YmvVICQw=
+github.com/gin-contrib/cors v1.7.2/go.mod h1:SUJVARKgQ40dmrzgXEVxj2m7Ig1v1qIboQkPDTQ9t2E=
+github.com/gin-contrib/gzip v1.0.1 h1:HQ8ENHODeLY7a4g1Au/46Z92bdGFl74OhxcZble9WJE=
+github.com/gin-contrib/gzip v1.0.1/go.mod h1:njt428fdUNRvjuJf16tZMYZ2Yl+WQB53X5wmhDwXvC4=
+github.com/gin-contrib/sessions v1.0.1 h1:3hsJyNs7v7N8OtelFmYXFrulAf6zSR7nW/putcPEHxI=
+github.com/gin-contrib/sessions v1.0.1/go.mod h1:ouxSFM24/OgIud5MJYQJLpy6AwxQ5EYO9yLhbtObGkM=
 github.com/gin-contrib/sse v0.1.0 h1:Y/yl/+YNO8GZSjAhjMsSuLt29uWRFHdHYUb5lYOV9qE=
 github.com/gin-contrib/sse v0.1.0/go.mod h1:RHrZQHXnP2xjPF+u1gW/2HnVO7nvIa9PG3Gm+fLHvGI=
-github.com/gin-contrib/static v1.1.1 h1:XEvBd4DDLG1HBlyPBQU1XO8NlTpw6mgdqcPteetYA5k=
-github.com/gin-contrib/static v1.1.1/go.mod h1:yRGmar7+JYvbMLRPIi4H5TVVSBwULfT9vetnVD0IO74=
-github.com/gin-gonic/gin v1.9.1 h1:4idEAncQnU5cB7BeOkPtxjfCSye0AAm1R0RVIqJ+Jmg=
-github.com/gin-gonic/gin v1.9.1/go.mod h1:hPrL7YrpYKXt5YId3A/Tnip5kqbEAP+KLuI3SUcPTeU=
+github.com/gin-contrib/static v1.1.2 h1:c3kT4bFkUJn2aoRU3s6XnMjJT8J6nNWJkR0NglqmlZ4=
+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=
 github.com/go-playground/universal-translator v0.18.1 h1:Bcnm0ZwsGyWbCzImXv+pAJnYK9S473LQFuzCbDbfSFY=
 github.com/go-playground/universal-translator v0.18.1/go.mod h1:xekY+UJKNuX9WP91TpwSH2VMlDf28Uj24BCp08ZFTUY=
-github.com/go-playground/validator/v10 v10.19.0 h1:ol+5Fu+cSq9JD7SoSqe04GMI92cbn0+wvQ3bZ8b/AU4=
-github.com/go-playground/validator/v10 v10.19.0/go.mod h1:dbuPbCMFw/DrkbEynArYaCwl3amGuJotoKCe95atGMM=
+github.com/go-playground/validator/v10 v10.20.0 h1:K9ISHbSaI0lyB2eWMPJo+kOS/FBExVwjEviJTixqxL8=
+github.com/go-playground/validator/v10 v10.20.0/go.mod h1:dbuPbCMFw/DrkbEynArYaCwl3amGuJotoKCe95atGMM=
 github.com/go-redis/redis/v8 v8.11.5 h1:AcZZR7igkdvfVmQTPnu9WE37LRrO/YrBH5zWyjDC0oI=
 github.com/go-redis/redis/v8 v8.11.5/go.mod h1:gREzHqY1hg6oD9ngVRbLStwAWKhA0FEgq8Jd4h5lpwo=
 github.com/go-sql-driver/mysql v1.7.0/go.mod h1:OXbVy3sEdcQ2Doequ6Z5BW6fXNQTmx+9S1MCJN5yJMI=
 github.com/go-sql-driver/mysql v1.8.1 h1:LedoTUt/eveggdHS9qUFC1EFSa8bU2+1pZjSRpvNJ1Y=
 github.com/go-sql-driver/mysql v1.8.1/go.mod h1:wEBSXgmK//2ZFJyE+qWnIsVGmvmEKlqwuVSjsCm7DZg=
-github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU=
-github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
+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/google/go-cmp v0.5.6 h1:BKbKCqvP6I+rmFHt06ZmyQtvB8xAkWdhFyr0ZUNZcxQ=
+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/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=
@@ -94,6 +146,8 @@ github.com/jinzhu/inflection v1.0.0 h1:K317FqzuhWc8YvSVlFMCCUb36O/S9MCKRDI7QkRKD
 github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkryuEj+Srlc=
 github.com/jinzhu/now v1.1.5 h1:/o9tlHleP7gOFmsnYNz3RGnqzefHA47wQpKrrdTIwXQ=
 github.com/jinzhu/now v1.1.5/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8=
+github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0=
+github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4=
 github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM=
 github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo=
 github.com/jtolds/gls v4.20.0+incompatible h1:xdiiI2gbIgH/gLH7ADydsJ1uDOEzR8yvV7C0MuV77Wo=
@@ -109,8 +163,8 @@ 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 v2.0.3+incompatible h1:gXHsfypPkaMZrKbD5209QV9jbUTJKjyR5WD3HYQSd+U=
-github.com/mattn/go-sqlite3 v2.0.3+incompatible/go.mod h1:FPy6KqzDD04eiIsT53CuJW3U88zkxoIYsOqkbpncsNc=
+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/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=
@@ -119,14 +173,17 @@ 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/pelletier/go-toml/v2 v2.2.1 h1:9TA9+T8+8CUCO2+WYnDLCgrYi9+omqKXyjDtosvtEhg=
-github.com/pelletier/go-toml/v2 v2.2.1/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs=
+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=
 github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
-github.com/pkoukk/tiktoken-go v0.1.6 h1:JF0TlJzhTbrI30wCvFuiw6FzP2+/bR+FIxUdgEAcUsw=
-github.com/pkoukk/tiktoken-go v0.1.6/go.mod h1:9NiV+i9mJKGj1rYOT+njbv+ZwA/zJxYdewGl6qVatpg=
+github.com/pkoukk/tiktoken-go v0.1.7 h1:qOBHXX4PHtvIvmOtyg1EeKlwFRiMKAcoMp4Q+bLQDmw=
+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=
@@ -148,26 +205,96 @@ 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 v0.7.0 h1:pskyeJh/3AmoQ8CPE95vxHLqp1G1GfGNXTmcl9NEKTc=
-golang.org/x/arch v0.7.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys=
-golang.org/x/crypto v0.22.0 h1:g1v0xeRhjcugydODzvb3mEM9SQ0HGp9s/nh3COQ/C30=
-golang.org/x/crypto v0.22.0/go.mod h1:vr6Su+7cTlO45qkww3VDJlzDn0ctJvRgYbC2NvXHt+M=
-golang.org/x/image v0.15.0 h1:kOELfmgrmJlw4Cdb7g/QGuB3CvDrXbqEIww/pNtNBm8=
-golang.org/x/image v0.15.0/go.mod h1:HUYqC05R2ZcZ3ejNQsIHQDQiwWM4JBqmm6MKANTp4LE=
-golang.org/x/net v0.24.0 h1:1PcaxkF854Fu3+lvBIx5SYn9wRlBzzcnHZSiaFFAb0w=
-golang.org/x/net v0.24.0/go.mod h1:2Q7sJY5mzlzWjKtYUEXSlBWCdyaioyXzRB2RtU8KVE8=
+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/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 v0.19.0 h1:q5f1RH2jigJ1MoAWp2KTp3gm5zAGFUTarQZ5U386+4o=
-golang.org/x/sys v0.19.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ=
-golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
-golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE=
-google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=
-google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
+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=
 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=
@@ -182,7 +309,9 @@ gorm.io/driver/postgres v1.5.7/go.mod h1:3e019WlBaYI5o5LIdNV+LyxCMNtLOQETBXL2h4c
 gorm.io/driver/sqlite v1.5.5 h1:7MDMtUZhV065SilG62E0MquljeArQZNfJnjd9i9gx3E=
 gorm.io/driver/sqlite v1.5.5/go.mod h1:6NgQ7sQWAIFsPrJJl1lSNSu2TABh0ZZ/zm5fosATavE=
 gorm.io/gorm v1.25.7/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8=
-gorm.io/gorm v1.25.9 h1:wct0gxZIELDk8+ZqF/MVnHLkA1rvYlBWUMv2EdsK1g8=
-gorm.io/gorm v1.25.9/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=

main.go

@@ -6,7 +6,9 @@ import (
 	"github.com/gin-contrib/sessions"
 	"github.com/gin-contrib/sessions/cookie"
 	"github.com/gin-gonic/gin"
+	_ "github.com/joho/godotenv/autoload"
 	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/client"
 	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/controller"
@@ -22,29 +24,22 @@ import (
 var buildFS embed.FS
 
 func main() {
+	common.Init()
 	logger.SetupLogger()
-	logger.SysLog(fmt.Sprintf("One API %s started", common.Version))
-	if os.Getenv("GIN_MODE") != "debug" {
+	logger.SysLogf("One API %s started", common.Version)
+
+	if os.Getenv("GIN_MODE") != gin.DebugMode {
 		gin.SetMode(gin.ReleaseMode)
 	}
 	if config.DebugEnabled {
 		logger.SysLog("running in debug mode")
 	}
-	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
-	}
+	model.InitDB()
+	model.InitLogDB()
+
+	var err error
 	err = model.CreateRootAccountIfNeed()
 	if err != nil {
 		logger.FatalLog("database init error: " + err.Error())
@@ -71,7 +66,7 @@ func main() {
 	}
 	if config.MemoryCacheEnabled {
 		logger.SysLog("memory cache enabled")
-		logger.SysError(fmt.Sprintf("sync frequency: %d seconds", config.SyncFrequency))
+		logger.SysLog(fmt.Sprintf("sync frequency: %d seconds", config.SyncFrequency))
 		model.InitChannelCache()
 	}
 	if config.MemoryCacheEnabled {
@@ -94,6 +89,7 @@ func main() {
 		logger.SysLog("metric enabled, will disable channel if too much request failed")
 	}
 	openai.InitTokenEncoders()
+	client.Init()
 
 	// Initialize HTTP server
 	server := gin.New()
@@ -111,6 +107,7 @@ func main() {
 	if port == "" {
 		port = strconv.Itoa(*common.Port)
 	}
+	logger.SysLogf("server started on http://localhost:%s", port)
 	err = server.Run(":" + port)
 	if err != nil {
 		logger.FatalLog("failed to start HTTP server: " + err.Error())

middleware/auth.go

@@ -5,6 +5,7 @@ import (
 	"github.com/gin-contrib/sessions"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/blacklist"
+	"github.com/songquanpeng/one-api/common/ctxkey"
 	"github.com/songquanpeng/one-api/common/network"
 	"github.com/songquanpeng/one-api/model"
 	"net/http"
@@ -120,25 +121,31 @@ func TokenAuth() func(c *gin.Context) {
 			abortWithMessage(c, http.StatusBadRequest, err.Error())
 			return
 		}
-		c.Set("request_model", requestModel)
+		c.Set(ctxkey.RequestModel, requestModel)
 		if token.Models != nil && *token.Models != "" {
-			c.Set("available_models", *token.Models)
+			c.Set(ctxkey.AvailableModels, *token.Models)
 			if requestModel != "" && !isModelInList(requestModel, *token.Models) {
 				abortWithMessage(c, http.StatusForbidden, fmt.Sprintf("该令牌无权使用模型：%s", requestModel))
 				return
 			}
 		}
-		c.Set("id", token.UserId)
-		c.Set("token_id", token.Id)
-		c.Set("token_name", token.Name)
+		c.Set(ctxkey.Id, token.UserId)
+		c.Set(ctxkey.TokenId, token.Id)
+		c.Set(ctxkey.TokenName, token.Name)
 		if len(parts) > 1 {
 			if model.IsAdmin(token.UserId) {
-				c.Set("specific_channel_id", parts[1])
+				c.Set(ctxkey.SpecificChannelId, parts[1])
 			} else {
 				abortWithMessage(c, http.StatusForbidden, "普通用户不支持指定渠道")
 				return
 			}
 		}
+
+		// set channel id for proxy relay
+		if channelId := c.Param("channelid"); channelId != "" {
+			c.Set(ctxkey.SpecificChannelId, channelId)
+		}
+
 		c.Next()
 	}
 }

middleware/distributor.go

@@ -3,7 +3,6 @@ package middleware
 import (
 	"fmt"
 	"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"
 	"github.com/songquanpeng/one-api/model"
@@ -13,17 +12,17 @@ import (
 )
 
 type ModelRequest struct {
-	Model string `json:"model"`
+	Model string `json:"model" form:"model"`
 }
 
 func Distribute() func(c *gin.Context) {
 	return func(c *gin.Context) {
-		userId := c.GetInt("id")
+		userId := c.GetInt(ctxkey.Id)
 		userGroup, _ := model.CacheGetUserGroup(userId)
-		c.Set("group", userGroup)
+		c.Set(ctxkey.Group, userGroup)
 		var requestModel string
 		var channel *model.Channel
-		channelId, ok := c.Get("specific_channel_id")
+		channelId, ok := c.Get(ctxkey.SpecificChannelId)
 		if ok {
 			id, err := strconv.Atoi(channelId.(string))
 			if err != nil {
@@ -40,7 +39,7 @@ func Distribute() func(c *gin.Context) {
 				return
 			}
 		} else {
-			requestModel = c.GetString("request_model")
+			requestModel = c.GetString(ctxkey.RequestModel)
 			var err error
 			channel, err = model.CacheGetRandomSatisfiedChannel(userGroup, requestModel, false)
 			if err != nil {
@@ -59,28 +58,38 @@ func Distribute() func(c *gin.Context) {
 }
 
 func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, modelName string) {
-	c.Set("channel", channel.Type)
-	c.Set("channel_id", channel.Id)
-	c.Set("channel_name", channel.Name)
-	c.Set("model_mapping", channel.GetModelMapping())
+	c.Set(ctxkey.Channel, channel.Type)
+	c.Set(ctxkey.ChannelId, channel.Id)
+	c.Set(ctxkey.ChannelName, channel.Name)
+	c.Set(ctxkey.ModelMapping, channel.GetModelMapping())
 	c.Set(ctxkey.OriginalModel, modelName) // for retry
 	c.Request.Header.Set("Authorization", fmt.Sprintf("Bearer %s", channel.Key))
-	c.Set("base_url", channel.GetBaseURL())
+	c.Set(ctxkey.BaseURL, channel.GetBaseURL())
+	cfg, _ := channel.LoadConfig()
 	// this is for backward compatibility
+	if channel.Other != nil {
 		switch channel.Type {
 		case channeltype.Azure:
-		c.Set(config.KeyAPIVersion, channel.Other)
+			if cfg.APIVersion == "" {
+				cfg.APIVersion = *channel.Other
+			}
 		case channeltype.Xunfei:
-		c.Set(config.KeyAPIVersion, channel.Other)
+			if cfg.APIVersion == "" {
+				cfg.APIVersion = *channel.Other
+			}
 		case channeltype.Gemini:
-		c.Set(config.KeyAPIVersion, channel.Other)
+			if cfg.APIVersion == "" {
+				cfg.APIVersion = *channel.Other
+			}
 		case channeltype.AIProxyLibrary:
-		c.Set(config.KeyLibraryID, channel.Other)
+			if cfg.LibraryID == "" {
+				cfg.LibraryID = *channel.Other
+			}
 		case channeltype.Ali:
-		c.Set(config.KeyPlugin, channel.Other)
-	}
-	cfg, _ := channel.LoadConfig()
-	for k, v := range cfg {
-		c.Set(config.KeyPrefix+k, v)
+			if cfg.Plugin == "" {
+				cfg.Plugin = *channel.Other
 			}
 		}
+	}
+	c.Set(ctxkey.Config, cfg)
+}

middleware/logger.go

@@ -3,14 +3,14 @@ package middleware
 import (
 	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/common/helper"
 )
 
 func SetUpLogger(server *gin.Engine) {
 	server.Use(gin.LoggerWithFormatter(func(param gin.LogFormatterParams) string {
 		var requestID string
 		if param.Keys != nil {
-			requestID = param.Keys[logger.RequestIdKey].(string)
+			requestID = param.Keys[helper.RequestIdKey].(string)
 		}
 		return fmt.Sprintf("[GIN] %s | %s | %3d | %13v | %15s | %7s %s\n",
 			param.TimeStamp.Format("2006/01/02 - 15:04:05"),

middleware/rate-limit.go

@@ -3,11 +3,12 @@ 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"
@@ -70,6 +71,11 @@ 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)

middleware/request-id.go

@@ -4,16 +4,15 @@ import (
 	"context"
 	"github.com/gin-gonic/gin"
 	"github.com/songquanpeng/one-api/common/helper"
-	"github.com/songquanpeng/one-api/common/logger"
 )
 
 func RequestId() func(c *gin.Context) {
 	return func(c *gin.Context) {
 		id := helper.GenRequestID()
-		c.Set(logger.RequestIdKey, id)
-		ctx := context.WithValue(c.Request.Context(), logger.RequestIdKey, id)
+		c.Set(helper.RequestIdKey, id)
+		ctx := context.WithValue(c.Request.Context(), helper.RequestIdKey, id)
 		c.Request = c.Request.WithContext(ctx)
-		c.Header(logger.RequestIdKey, id)
+		c.Header(helper.RequestIdKey, id)
 		c.Next()
 	}
 }

middleware/utils.go

@@ -12,7 +12,7 @@ import (
 func abortWithMessage(c *gin.Context, statusCode int, message string) {
 	c.JSON(statusCode, gin.H{
 		"error": gin.H{
-			"message": helper.MessageWithRequestId(message, c.GetString(logger.RequestIdKey)),
+			"message": helper.MessageWithRequestId(message, c.GetString(helper.RequestIdKey)),
 			"type":    "one_api_error",
 		},
 	})

model/channel.go

@@ -3,6 +3,7 @@ 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"
@@ -27,7 +28,7 @@ type Channel struct {
 	TestTime           int64   `json:"test_time" gorm:"bigint"`
 	ResponseTime       int     `json:"response_time"` // in milliseconds
 	BaseURL            *string `json:"base_url" gorm:"column:base_url;default:''"`
-	Other              string  `json:"other"`   // DEPRECATED: please save config to field Config
+	Other              *string `json:"other"`   // DEPRECATED: please save config to field Config
 	Balance            float64 `json:"balance"` // in USD
 	BalanceUpdatedTime int64   `json:"balance_updated_time" gorm:"bigint"`
 	Models             string  `json:"models"`
@@ -38,6 +39,18 @@ type Channel struct {
 	Config             string  `json:"config"`
 }
 
+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"`
+}
+
 func GetAllChannels(startIdx int, num int, scope string) ([]*Channel, error) {
 	var channels []*Channel
 	var err error
@@ -161,14 +174,14 @@ func (channel *Channel) Delete() error {
 	return err
 }
 
-func (channel *Channel) LoadConfig() (map[string]string, error) {
+func (channel *Channel) LoadConfig() (ChannelConfig, error) {
+	var cfg ChannelConfig
 	if channel.Config == "" {
-		return nil, nil
+		return cfg, nil
 	}
-	cfg := make(map[string]string)
 	err := json.Unmarshal([]byte(channel.Config), &cfg)
 	if err != nil {
-		return nil, err
+		return cfg, err
 	}
 	return cfg, nil
 }

model/log.go

@@ -3,6 +3,7 @@ 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"
@@ -152,7 +153,11 @@ 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) {
-	tx := LOG_DB.Table("logs").Select("ifnull(sum(quota),0)")
+	ifnull := "ifnull"
+	if common.UsingPostgreSQL {
+		ifnull = "COALESCE"
+	}
+	tx := LOG_DB.Table("logs").Select(fmt.Sprintf("%s(sum(quota),0)", ifnull))
 	if username != "" {
 		tx = tx.Where("username = ?", username)
 	}
@@ -176,7 +181,11 @@ 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) {
-	tx := LOG_DB.Table("logs").Select("ifnull(sum(prompt_tokens),0) + ifnull(sum(completion_tokens),0)")
+	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))
 	if username != "" {
 		tx = tx.Where("username = ?", username)
 	}

model/main.go

@@ -1,6 +1,7 @@
 package model
 
 import (
+	"database/sql"
 	"fmt"
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/config"
@@ -29,13 +30,17 @@ 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: random.GetUUID(),
+			AccessToken: accessToken,
 			Quota:       500000000000000,
 		}
 		DB.Create(&rootUser)
@@ -60,10 +65,22 @@ func CreateRootAccountIfNeed() error {
 }
 
 func chooseDB(envName string) (*gorm.DB, error) {
-	if os.Getenv(envName) != "" {
 	dsn := os.Getenv(envName)
-		if strings.HasPrefix(dsn, "postgres://") {
+
+	switch {
+	case strings.HasPrefix(dsn, "postgres://"):
 		// Use PostgreSQL
+		return openPostgreSQL(dsn)
+	case dsn != "":
+		// Use MySQL
+		return openMySQL(dsn)
+	default:
+		// Use SQLite
+		return openSQLite()
+	}
+}
+
+func openPostgreSQL(dsn string) (*gorm.DB, error) {
 	logger.SysLog("using PostgreSQL as database")
 	common.UsingPostgreSQL = true
 	return gorm.Open(postgres.New(postgres.Config{
@@ -73,77 +90,131 @@ func chooseDB(envName string) (*gorm.DB, error) {
 		PrepareStmt: true, // precompile SQL
 	})
 }
-		// Use MySQL
+
+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
 	})
 }
-	// Use SQLite
+
+func openSQLite() (*gorm.DB, error) {
 	logger.SysLog("SQL_DSN not set, using SQLite as database")
 	common.UsingSQLite = true
-	config := fmt.Sprintf("?_busy_timeout=%d", common.SQLiteBusyTimeout)
-	return gorm.Open(sqlite.Open(common.SQLitePath+config), &gorm.Config{
+	dsn := fmt.Sprintf("%s?_busy_timeout=%d", common.SQLitePath, common.SQLiteBusyTimeout)
+	return gorm.Open(sqlite.Open(dsn), &gorm.Config{
 		PrepareStmt: true, // precompile SQL
 	})
 }
 
-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()
+func InitDB() {
+	var err error
+	DB, err = chooseDB("SQL_DSN")
 	if err != nil {
-			return nil, err
+		logger.FatalLog("failed to initialize database: " + err.Error())
+		return
 	}
-		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 db, err
+		return
 	}
+
 	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
+	if err = migrateDB(); err != nil {
+		logger.FatalLog("failed to migrate database: " + err.Error())
+		return
 	}
 	logger.SysLog("database migrated")
-		return db, err
-	} else {
-		logger.FatalLog(err)
 }
-	return db, err
+
+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
 }
 
 func closeDB(db *gorm.DB) error {

model/option.go

@@ -28,6 +28,7 @@ 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)
@@ -130,6 +131,8 @@ 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":
@@ -176,6 +179,18 @@ 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":

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:"default:''"`           // allowed models
+	Models         *string `json:"models" gorm:"type:text"`            // allowed models
 	Subnet         *string `json:"subnet" gorm:"default:''"`           // allowed subnet
 }
 
@@ -121,30 +121,40 @@ func GetTokenById(id int) (*Token, error) {
 	return &token, err
 }
 
-func (token *Token) Insert() error {
+func (t *Token) Insert() error {
 	var err error
-	err = DB.Create(token).Error
+	err = DB.Create(t).Error
 	return err
 }
 
 // Update Make sure your token's fields is completed, because this will update non-zero values
-func (token *Token) Update() error {
+func (t *Token) Update() error {
 	var err error
-	err = DB.Model(token).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota", "models", "subnet").Updates(token).Error
+	err = DB.Model(t).Select("name", "status", "expired_time", "remain_quota", "unlimited_quota", "models", "subnet").Updates(t).Error
 	return err
 }
 
-func (token *Token) SelectUpdate() error {
+func (t *Token) SelectUpdate() error {
 	// This can update zero values
-	return DB.Model(token).Select("accessed_time", "status").Updates(token).Error
+	return DB.Model(t).Select("accessed_time", "status").Updates(t).Error
 }
 
-func (token *Token) Delete() error {
+func (t *Token) Delete() error {
 	var err error
-	err = DB.Delete(token).Error
+	err = DB.Delete(t).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 {
@@ -254,14 +264,14 @@ 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)

model/user.go

@@ -6,6 +6,7 @@ import (
 	"github.com/songquanpeng/one-api/common"
 	"github.com/songquanpeng/one-api/common/blacklist"
 	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/common/helper"
 	"github.com/songquanpeng/one-api/common/logger"
 	"github.com/songquanpeng/one-api/common/random"
 	"gorm.io/gorm"
@@ -38,6 +39,7 @@ 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"`
@@ -140,6 +142,22 @@ func (user *User) Insert(inviterId int) error {
 			RecordLog(inviterId, LogTypeSystem, fmt.Sprintf("邀请用户赠送 %s", common.LogQuota(config.QuotaForInviter)))
 		}
 	}
+	// create default token
+	cleanToken := Token{
+		UserId:         user.Id,
+		Name:           "default",
+		Key:            random.GenerateKey(),
+		CreatedTime:    helper.GetTimestamp(),
+		AccessedTime:   helper.GetTimestamp(),
+		ExpiredTime:    -1,
+		RemainQuota:    -1,
+		UnlimitedQuota: true,
+	}
+	result.Error = cleanToken.Insert()
+	if result.Error != nil {
+		// do not block
+		logger.SysError(fmt.Sprintf("create default token for user %d failed: %s", user.Id, result.Error.Error()))
+	}
 	return nil
 }
 
@@ -228,6 +246,14 @@ 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！")
@@ -260,6 +286,10 @@ 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
 }

monitor/manage.go

@@ -1,10 +1,11 @@
 package monitor
 
 import (
-	"github.com/songquanpeng/one-api/common/config"
-	"github.com/songquanpeng/one-api/relay/model"
 	"net/http"
 	"strings"
+
+	"github.com/songquanpeng/one-api/common/config"
+	"github.com/songquanpeng/one-api/relay/model"
 )
 
 func ShouldDisableChannel(err *model.Error, statusCode int) bool {
@@ -18,31 +19,23 @@ func ShouldDisableChannel(err *model.Error, statusCode int) bool {
 		return true
 	}
 	switch err.Type {
-	case "insufficient_quota":
-		return true
-	// https://docs.anthropic.com/claude/reference/errors
-	case "authentication_error":
-		return true
-	case "permission_error":
-		return true
-	case "forbidden":
+	case "insufficient_quota", "authentication_error", "permission_error", "forbidden":
 		return true
 	}
 	if err.Code == "invalid_api_key" || err.Code == "account_deactivated" {
 		return true
 	}
-	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") {
+
+	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") {
 		return true
 	}
 	return false

relay/adaptor.go

@@ -7,11 +7,17 @@ import (
 	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
 	"github.com/songquanpeng/one-api/relay/adaptor/aws"
 	"github.com/songquanpeng/one-api/relay/adaptor/baidu"
+	"github.com/songquanpeng/one-api/relay/adaptor/cloudflare"
+	"github.com/songquanpeng/one-api/relay/adaptor/cohere"
+	"github.com/songquanpeng/one-api/relay/adaptor/coze"
+	"github.com/songquanpeng/one-api/relay/adaptor/deepl"
 	"github.com/songquanpeng/one-api/relay/adaptor/gemini"
 	"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"
@@ -43,6 +49,18 @@ func GetAdaptor(apiType int) adaptor.Adaptor {
 		return &zhipu.Adaptor{}
 	case apitype.Ollama:
 		return &ollama.Adaptor{}
+	case apitype.Coze:
+		return &coze.Adaptor{}
+	case apitype.Cohere:
+		return &cohere.Adaptor{}
+	case apitype.Cloudflare:
+		return &cloudflare.Adaptor{}
+	case apitype.DeepL:
+		return &deepl.Adaptor{}
+	case apitype.VertexAI:
+		return &vertexai.Adaptor{}
+	case apitype.Proxy:
+		return &proxy.Adaptor{}
 	}
 	return nil
 }

relay/adaptor/aiproxy/adaptor.go

@@ -4,7 +4,6 @@ import (
 	"errors"
 	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
@@ -13,10 +12,11 @@ import (
 )
 
 type Adaptor struct {
+	meta *meta.Meta
 }
 
 func (a *Adaptor) Init(meta *meta.Meta) {
-
+	a.meta = meta
 }
 
 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
@@ -34,7 +34,7 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 		return nil, errors.New("request is nil")
 	}
 	aiProxyLibraryRequest := ConvertRequest(*request)
-	aiProxyLibraryRequest.LibraryId = c.GetString(config.KeyLibraryID)
+	aiProxyLibraryRequest.LibraryId = a.meta.Config.LibraryID
 	return aiProxyLibraryRequest, nil
 }
 

relay/adaptor/aiproxy/main.go

@@ -4,6 +4,12 @@ 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"
@@ -12,10 +18,6 @@ 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
@@ -89,6 +91,7 @@ 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 {
@@ -102,60 +105,48 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-	dataChan := make(chan string)
-	stopChan := make(chan bool)
-	go func() {
+
+	common.SetEventStreamHeaders(c)
+
 	for scanner.Scan() {
 		data := scanner.Text()
-			if len(data) < 5 { // ignore blank line or wrong format
-				continue
-			}
-			if data[:5] != "data:" {
+		if len(data) < 5 || data[:5] != "data:" {
 			continue
 		}
 		data = data[5:]
-			dataChan <- data
-		}
-		stopChan <- true
-	}()
-	common.SetEventStreamHeaders(c)
-	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
+			continue
 		}
 		if len(AIProxyLibraryResponse.Documents) != 0 {
 			documents = AIProxyLibraryResponse.Documents
 		}
 		response := streamResponseAIProxyLibrary2OpenAI(&AIProxyLibraryResponse)
-			jsonResponse, err := json.Marshal(response)
+		err = render.ObjectData(c, response)
 		if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
+			logger.SysError(err.Error())
 		}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
-			return true
-		case <-stopChan:
+	}
+
+	if err := scanner.Err(); err != nil {
+		logger.SysError("error reading stream: " + err.Error())
+	}
+
 	response := documentsAIProxyLibrary(documents)
-			jsonResponse, err := json.Marshal(response)
+	err := render.ObjectData(c, response)
 	if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
+		logger.SysError(err.Error())
 	}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
-			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
-			return false
-		}
-	})
-	err := resp.Body.Close()
+	render.Done(c)
+
+	err = resp.Body.Close()
 	if err != nil {
 		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
 	}
+
 	return nil, &usage
 }
 

relay/adaptor/ali/adaptor.go

@@ -4,7 +4,6 @@ import (
 	"errors"
 	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/config"
 	"github.com/songquanpeng/one-api/relay/adaptor"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
@@ -16,10 +15,11 @@ import (
 // https://help.aliyun.com/zh/dashscope/developer-reference/api-details
 
 type Adaptor struct {
+	meta *meta.Meta
 }
 
 func (a *Adaptor) Init(meta *meta.Meta) {
-
+	a.meta = meta
 }
 
 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
@@ -47,8 +47,8 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *me
 	if meta.Mode == relaymode.ImagesGenerations {
 		req.Header.Set("X-DashScope-Async", "enable")
 	}
-	if c.GetString(config.KeyPlugin) != "" {
-		req.Header.Set("X-DashScope-Plugin", c.GetString(config.KeyPlugin))
+	if a.meta.Config.Plugin != "" {
+		req.Header.Set("X-DashScope-Plugin", a.meta.Config.Plugin)
 	}
 	return nil
 }

relay/adaptor/ali/main.go

@@ -3,15 +3,18 @@ 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
@@ -57,7 +60,7 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
 
 func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
 	return &EmbeddingRequest{
-		Model: "text-embedding-v1",
+		Model: request.Model,
 		Input: struct {
 			Texts []string `json:"texts"`
 		}{
@@ -100,8 +103,9 @@ 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
@@ -181,32 +185,21 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		return 0, nil, nil
 	})
-	dataChan := make(chan string)
-	stopChan := make(chan bool)
-	go func() {
+
+	common.SetEventStreamHeaders(c)
+
 	for scanner.Scan() {
 		data := scanner.Text()
-			if len(data) < 5 { // ignore blank line or wrong format
-				continue
-			}
-			if data[:5] != "data:" {
+		if len(data) < 5 || data[:5] != "data:" {
 			continue
 		}
 		data = data[5:]
-			dataChan <- data
-		}
-		stopChan <- true
-	}()
-	common.SetEventStreamHeaders(c)
-	//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
+			continue
 		}
 		if aliResponse.Usage.OutputTokens != 0 {
 			usage.PromptTokens = aliResponse.Usage.InputTokens
@@ -215,22 +208,20 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 		}
 		response := streamResponseAli2OpenAI(&aliResponse)
 		if response == nil {
-				return true
+			continue
 		}
-			//response.Choices[0].Delta.Content = strings.TrimPrefix(response.Choices[0].Delta.Content, lastResponseText)
-			//lastResponseText = aliResponse.Output.Text
-			jsonResponse, err := json.Marshal(response)
+		err = render.ObjectData(c, response)
 		if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
+			logger.SysError(err.Error())
 		}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
-			return true
-		case <-stopChan:
-			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
-			return false
 	}
-	})
+
+	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

relay/adaptor/anthropic/adaptor.go

@@ -3,12 +3,14 @@ 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 {
@@ -31,6 +33,13 @@ 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
 }
 

relay/adaptor/anthropic/constants.go

@@ -5,4 +5,5 @@ var ModelList = []string{
 	"claude-3-haiku-20240307",
 	"claude-3-sonnet-20240229",
 	"claude-3-opus-20240229",
+	"claude-3-5-sonnet-20240620",
 }

relay/adaptor/anthropic/main.go

@@ -4,6 +4,11 @@ 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/helper"
@@ -11,9 +16,6 @@ import (
 	"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"
 )
 
 func stopReasonClaude2OpenAI(reason *string) string {
@@ -27,12 +29,30 @@ 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,
@@ -40,6 +60,24 @@ 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
@@ -62,7 +100,24 @@ 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
 		}
@@ -95,16 +150,35 @@ 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 {
@@ -118,6 +192,10 @@ 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" {
@@ -134,12 +212,27 @@ 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,
 		},
 		FinishReason: stopReasonClaude2OpenAI(claudeResponse.StopReason),
 	}
@@ -168,64 +261,77 @@ 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
-	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 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())
-				return true
+			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.
 				modelName = meta.Model
 				id = fmt.Sprintf("chatcmpl-%s", meta.Id)
-				return true
+				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
+					}
+				}
+			}
 		}
 		if response == nil {
-				return true
+			continue
 		}
+
 		response.Id = id
 		response.Model = modelName
 		response.Created = createdTime
-			jsonStr, err := json.Marshal(response)
+
+		for _, choice := range response.Choices {
+			if len(choice.Delta.ToolCalls) > 0 {
+				lastToolCallChoice = choice
+			}
+		}
+		err = render.ObjectData(c, response)
 		if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
+			logger.SysError(err.Error())
 		}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)})
-			return true
-		case <-stopChan:
-			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
-			return false
 	}
-	})
-	_ = resp.Body.Close()
+
+	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
 }
 

relay/adaptor/anthropic/model.go

@@ -16,6 +16,12 @@ 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 {
@@ -23,6 +29,18 @@ 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"`
@@ -33,6 +51,8 @@ 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"`
 }
 
@@ -61,6 +81,7 @@ 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"`
 }

relay/adaptor/aws/adaptor.go

@@ -0,0 +1,84 @@
+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/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/relay/adaptor"
+	"github.com/songquanpeng/one-api/relay/adaptor/aws/utils"
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/model"
+)
+
+var _ adaptor.Adaptor = new(Adaptor)
+
+type Adaptor struct {
+	awsAdapter utils.AwsAdapter
+
+	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) 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
+}
+
+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
+}

relay/adaptor/aws/claude/adapter.go

@@ -0,0 +1,37 @@
+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
+}

relay/adaptor/aws/claude/main.go

@@ -5,73 +5,48 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"github.com/songquanpeng/one-api/common/config"
-	"github.com/songquanpeng/one-api/common/ctxkey"
 	"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/aws/aws-sdk-go-v2/service/bedrockruntime/types"
 	"github.com/gin-gonic/gin"
 	"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 newAwsClient(c *gin.Context) (*bedrockruntime.Client, error) {
-	ak := c.GetString(config.KeyAK)
-	sk := c.GetString(config.KeySK)
-	region := c.GetString(config.KeyRegion)
-	client := bedrockruntime.New(bedrockruntime.Options{
-		Region:      region,
-		Credentials: aws.NewCredentialsCache(credentials.NewStaticCredentialsProvider(ak, sk, "")),
-	})
-
-	return client, nil
-}
-
-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{
+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",
 }
 
 func awsModelID(requestModel string) (string, error) {
-	if awsModelID, ok := awsModelIDMap[requestModel]; ok {
+	if awsModelID, ok := AwsModelIDMap[requestModel]; ok {
 		return awsModelID, nil
 	}
 
 	return "", errors.Errorf("model %s not found", requestModel)
 }
 
-func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
-	awsCli, err := newAwsClient(c)
-	if err != nil {
-		return wrapErr(errors.Wrap(err, "newAwsClient")), nil
-	}
-
+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 wrapErr(errors.Wrap(err, "awsModelID")), nil
+		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
 	}
 
 	awsReq := &bedrockruntime.InvokeModelInput{
@@ -82,30 +57,30 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 
 	claudeReq_, ok := c.Get(ctxkey.ConvertedRequest)
 	if !ok {
-		return wrapErr(errors.New("request not found")), nil
+		return utils.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 wrapErr(errors.Wrap(err, "copy request")), nil
+		return utils.WrapErr(errors.Wrap(err, "copy request")), nil
 	}
 
 	awsReq.Body, err = json.Marshal(awsClaudeReq)
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "marshal request")), nil
+		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
 	}
 
 	awsResp, err := awsCli.InvokeModel(c.Request.Context(), awsReq)
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "InvokeModel")), nil
+		return utils.WrapErr(errors.Wrap(err, "InvokeModel")), nil
 	}
 
 	claudeResponse := new(anthropic.Response)
 	err = json.Unmarshal(awsResp.Body, claudeResponse)
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "unmarshal response")), nil
+		return utils.WrapErr(errors.Wrap(err, "unmarshal response")), nil
 	}
 
 	openaiResp := anthropic.ResponseClaude2OpenAI(claudeResponse)
@@ -121,16 +96,11 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st
 	return nil, &usage
 }
 
-func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
+func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.ErrorWithStatusCode, *relaymodel.Usage) {
 	createdTime := helper.GetTimestamp()
-	awsCli, err := newAwsClient(c)
-	if err != nil {
-		return wrapErr(errors.Wrap(err, "newAwsClient")), nil
-	}
-
 	awsModelId, err := awsModelID(c.GetString(ctxkey.RequestModel))
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "awsModelID")), nil
+		return utils.WrapErr(errors.Wrap(err, "awsModelID")), nil
 	}
 
 	awsReq := &bedrockruntime.InvokeModelWithResponseStreamInput{
@@ -141,7 +111,7 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithSt
 
 	claudeReq_, ok := c.Get(ctxkey.ConvertedRequest)
 	if !ok {
-		return wrapErr(errors.New("request not found")), nil
+		return utils.WrapErr(errors.New("request not found")), nil
 	}
 	claudeReq := claudeReq_.(*anthropic.Request)
 
@@ -149,16 +119,16 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithSt
 		AnthropicVersion: "bedrock-2023-05-31",
 	}
 	if err = copier.Copy(awsClaudeReq, claudeReq); err != nil {
-		return wrapErr(errors.Wrap(err, "copy request")), nil
+		return utils.WrapErr(errors.Wrap(err, "copy request")), nil
 	}
 	awsReq.Body, err = json.Marshal(awsClaudeReq)
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "marshal request")), nil
+		return utils.WrapErr(errors.Wrap(err, "marshal request")), nil
 	}
 
 	awsResp, err := awsCli.InvokeModelWithResponseStream(c.Request.Context(), awsReq)
 	if err != nil {
-		return wrapErr(errors.Wrap(err, "InvokeModelWithResponseStream")), nil
+		return utils.WrapErr(errors.Wrap(err, "InvokeModelWithResponseStream")), nil
 	}
 	stream := awsResp.GetStream()
 	defer stream.Close()
@@ -166,6 +136,8 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithSt
 	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 {
@@ -186,8 +158,19 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithSt
 			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
+						}
+					}
+				}
 			}
 			if response == nil {
 				return true
@@ -195,6 +178,12 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*relaymodel.ErrorWithSt
 			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())

relay/adaptor/aws/claude/model.go

@@ -9,9 +9,12 @@ 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"`
 }

relay/adaptor/aws/llama3/adapter.go

@@ -0,0 +1,37 @@
+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
+}

relay/adaptor/aws/llama3/main.go

@@ -0,0 +1,231 @@
+// 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
+}

relay/adaptor/aws/llama3/main_test.go

@@ -0,0 +1,45 @@
+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)
+}

relay/adaptor/aws/llama3/model.go

@@ -0,0 +1,29 @@
+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"`
+}

relay/adaptor/aws/registry.go

@@ -0,0 +1,39 @@
+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
+	}
+}

relay/adaptor/aws/utils/adaptor.go

@@ -0,0 +1,51 @@
+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
+}

relay/adaptor/aws/utils/utils.go

@@ -0,0 +1,16 @@
+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(),
+		},
+	}
+}

relay/adaptor/azure/helper.go

@@ -1,15 +0,0 @@
-package azure
-
-import (
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common/config"
-)
-
-func GetAPIVersion(c *gin.Context) string {
-	query := c.Request.URL.Query()
-	apiVersion := query.Get("api-version")
-	if apiVersion == "" {
-		apiVersion = c.GetString(config.KeyAPIVersion)
-	}
-	return apiVersion
-}

relay/adaptor/baidu/main.go

@@ -5,18 +5,20 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/common"
-	"github.com/songquanpeng/one-api/common/logger"
-	"github.com/songquanpeng/one-api/relay/adaptor/openai"
-	"github.com/songquanpeng/one-api/relay/client"
-	"github.com/songquanpeng/one-api/relay/constant"
-	"github.com/songquanpeng/one-api/relay/model"
+	"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"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/model"
 )
 
 // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/flfmc9do2
@@ -137,40 +139,22 @@ 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(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() {
+	scanner.Split(bufio.ScanLines)
+
+	common.SetEventStreamHeaders(c)
+
 	for scanner.Scan() {
 		data := scanner.Text()
-			if len(data) < 6 { // ignore blank line or wrong format
+		if len(data) < 6 {
 			continue
 		}
 		data = data[6:]
-			dataChan <- data
-		}
-		stopChan <- true
-	}()
-	common.SetEventStreamHeaders(c)
-	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
+			continue
 		}
 		if baiduResponse.Usage.TotalTokens != 0 {
 			usage.TotalTokens = baiduResponse.Usage.TotalTokens
@@ -178,18 +162,18 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC
 			usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens
 		}
 		response := streamResponseBaidu2OpenAI(&baiduResponse)
-			jsonResponse, err := json.Marshal(response)
+		err = render.ObjectData(c, response)
 		if err != nil {
-				logger.SysError("error marshalling stream response: " + err.Error())
-				return true
+			logger.SysError(err.Error())
 		}
-			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)})
-			return true
-		case <-stopChan:
-			c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
-			return false
 	}
-	})
+
+	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

relay/adaptor/cloudflare/adaptor.go

@@ -0,0 +1,100 @@
+package cloudflare
+
+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"
+	"github.com/songquanpeng/one-api/relay/relaymode"
+)
+
+type Adaptor struct {
+	meta *meta.Meta
+}
+
+// ConvertImageRequest implements adaptor.Adaptor.
+func (*Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	return nil, errors.New("not implemented")
+}
+
+// ConvertImageRequest implements adaptor.Adaptor.
+
+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
+	}
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
+	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")
+	}
+	switch relayMode {
+	case relaymode.Completions:
+		return ConvertCompletionsRequest(*request), nil
+	case relaymode.ChatCompletions, relaymode.Embeddings:
+		return request, nil
+	default:
+		return nil, errors.New("not implemented")
+	}
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	return adaptor.DoRequestHelper(a, c, meta, requestBody)
+}
+
+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, resp, meta.PromptTokens, meta.ActualModelName)
+	} else {
+		err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "cloudflare"
+}

relay/adaptor/cloudflare/constant.go

@@ -0,0 +1,37 @@
+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",
+	"@hf/thebloke/deepseek-coder-6.7b-base-awq",
+	"@hf/thebloke/deepseek-coder-6.7b-instruct-awq",
+	"@cf/deepseek-ai/deepseek-math-7b-base",
+	"@cf/deepseek-ai/deepseek-math-7b-instruct",
+	"@cf/thebloke/discolm-german-7b-v1-awq",
+	"@cf/tiiuae/falcon-7b-instruct",
+	"@cf/google/gemma-2b-it-lora",
+	"@hf/google/gemma-7b-it",
+	"@cf/google/gemma-7b-it-lora",
+	"@hf/nousresearch/hermes-2-pro-mistral-7b",
+	"@hf/thebloke/llama-2-13b-chat-awq",
+	"@cf/meta-llama/llama-2-7b-chat-hf-lora",
+	"@cf/meta/llama-3-8b-instruct",
+	"@hf/thebloke/llamaguard-7b-awq",
+	"@hf/thebloke/mistral-7b-instruct-v0.1-awq",
+	"@hf/mistralai/mistral-7b-instruct-v0.2",
+	"@cf/mistral/mistral-7b-instruct-v0.2-lora",
+	"@hf/thebloke/neural-chat-7b-v3-1-awq",
+	"@cf/openchat/openchat-3.5-0106",
+	"@hf/thebloke/openhermes-2.5-mistral-7b-awq",
+	"@cf/microsoft/phi-2",
+	"@cf/qwen/qwen1.5-0.5b-chat",
+	"@cf/qwen/qwen1.5-1.8b-chat",
+	"@cf/qwen/qwen1.5-14b-chat-awq",
+	"@cf/qwen/qwen1.5-7b-chat-awq",
+	"@cf/defog/sqlcoder-7b-2",
+	"@hf/nexusflow/starling-lm-7b-beta",
+	"@cf/tinyllama/tinyllama-1.1b-chat-v1.0",
+	"@hf/thebloke/zephyr-7b-beta-awq",
+}

relay/adaptor/cloudflare/main.go

@@ -0,0 +1,115 @@
+package cloudflare
+
+import (
+	"bufio"
+	"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"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"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 StreamHandler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *model.Usage) {
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(bufio.ScanLines)
+
+	common.SetEventStreamHeaders(c)
+	id := helper.GetResponseID(c)
+	responseModel := c.GetString(ctxkey.OriginalModel)
+	var responseText string
+
+	for scanner.Scan() {
+		data := scanner.Text()
+		if len(data) < len("data: ") {
+			continue
+		}
+		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
+	}
+
+	usage := openai.ResponseText2Usage(responseText, responseModel, promptTokens)
+	return nil, usage
+}
+
+func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *model.Usage) {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	var response openai.TextResponse
+	err = json.Unmarshal(responseBody, &response)
+	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)
+	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)
+	return nil, usage
+}

relay/adaptor/cloudflare/model.go

@@ -0,0 +1,13 @@
+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"`
+}

relay/adaptor/cohere/adaptor.go

@@ -0,0 +1,64 @@
+package cohere
+
+import (
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+
+	"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"
+)
+
+type Adaptor struct{}
+
+// ConvertImageRequest implements adaptor.Adaptor.
+func (*Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	return nil, errors.New("not implemented")
+}
+
+// ConvertImageRequest implements adaptor.Adaptor.
+
+func (a *Adaptor) Init(meta *meta.Meta) {
+
+}
+
+func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
+	return fmt.Sprintf("%s/v1/chat", meta.BaseURL), nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
+	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")
+	}
+	return ConvertRequest(*request), nil
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	return adaptor.DoRequestHelper(a, c, meta, requestBody)
+}
+
+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, resp)
+	} else {
+		err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "Cohere"
+}

relay/adaptor/cohere/constant.go

@@ -0,0 +1,14 @@
+package cohere
+
+var ModelList = []string{
+	"command", "command-nightly",
+	"command-light", "command-light-nightly",
+	"command-r", "command-r-plus",
+}
+
+func init() {
+	num := len(ModelList)
+	for i := 0; i < num; i++ {
+		ModelList = append(ModelList, ModelList[i]+"-internet")
+	}
+}

relay/adaptor/cohere/main.go

@@ -0,0 +1,228 @@
+package cohere
+
+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/helper"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/model"
+)
+
+var (
+	WebSearchConnector = Connector{ID: "web-search"}
+)
+
+func stopReasonCohere2OpenAI(reason *string) string {
+	if reason == nil {
+		return ""
+	}
+	switch *reason {
+	case "COMPLETE":
+		return "stop"
+	default:
+		return *reason
+	}
+}
+
+func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
+	cohereRequest := Request{
+		Model:            textRequest.Model,
+		Message:          "",
+		MaxTokens:        textRequest.MaxTokens,
+		Temperature:      textRequest.Temperature,
+		P:                textRequest.TopP,
+		K:                textRequest.TopK,
+		Stream:           textRequest.Stream,
+		FrequencyPenalty: textRequest.FrequencyPenalty,
+		PresencePenalty:  textRequest.FrequencyPenalty,
+		Seed:             int(textRequest.Seed),
+	}
+	if cohereRequest.Model == "" {
+		cohereRequest.Model = "command-r"
+	}
+	if strings.HasSuffix(cohereRequest.Model, "-internet") {
+		cohereRequest.Model = strings.TrimSuffix(cohereRequest.Model, "-internet")
+		cohereRequest.Connectors = append(cohereRequest.Connectors, WebSearchConnector)
+	}
+	for _, message := range textRequest.Messages {
+		if message.Role == "user" {
+			cohereRequest.Message = message.Content.(string)
+		} else {
+			var role string
+			if message.Role == "assistant" {
+				role = "CHATBOT"
+			} else if message.Role == "system" {
+				role = "SYSTEM"
+			} else {
+				role = "USER"
+			}
+			cohereRequest.ChatHistory = append(cohereRequest.ChatHistory, ChatMessage{
+				Role:    role,
+				Message: message.Content.(string),
+			})
+		}
+	}
+	return &cohereRequest
+}
+
+func StreamResponseCohere2OpenAI(cohereResponse *StreamResponse) (*openai.ChatCompletionsStreamResponse, *Response) {
+	var response *Response
+	var responseText string
+	var finishReason string
+
+	switch cohereResponse.EventType {
+	case "stream-start":
+		return nil, nil
+	case "text-generation":
+		responseText += cohereResponse.Text
+	case "stream-end":
+		usage := cohereResponse.Response.Meta.Tokens
+		response = &Response{
+			Meta: Meta{
+				Tokens: Usage{
+					InputTokens:  usage.InputTokens,
+					OutputTokens: usage.OutputTokens,
+				},
+			},
+		}
+		finishReason = *cohereResponse.Response.FinishReason
+	default:
+		return nil, nil
+	}
+
+	var choice openai.ChatCompletionsStreamResponseChoice
+	choice.Delta.Content = responseText
+	choice.Delta.Role = "assistant"
+	if finishReason != "" {
+		choice.FinishReason = &finishReason
+	}
+	var openaiResponse openai.ChatCompletionsStreamResponse
+	openaiResponse.Object = "chat.completion.chunk"
+	openaiResponse.Choices = []openai.ChatCompletionsStreamResponseChoice{choice}
+	return &openaiResponse, response
+}
+
+func ResponseCohere2OpenAI(cohereResponse *Response) *openai.TextResponse {
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    "assistant",
+			Content: cohereResponse.Text,
+			Name:    nil,
+		},
+		FinishReason: stopReasonCohere2OpenAI(cohereResponse.FinishReason),
+	}
+	fullTextResponse := openai.TextResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", cohereResponse.ResponseID),
+		Model:   "model",
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+	}
+	return &fullTextResponse
+}
+
+func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	createdTime := helper.GetTimestamp()
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(bufio.ScanLines)
+
+	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
+		}
+
+		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
+	}
+
+	return nil, &usage
+}
+
+func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *model.Usage) {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	var cohereResponse Response
+	err = json.Unmarshal(responseBody, &cohereResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	if cohereResponse.ResponseID == "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: cohereResponse.Message,
+				Type:    cohereResponse.Message,
+				Param:   "",
+				Code:    resp.StatusCode,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	fullTextResponse := ResponseCohere2OpenAI(&cohereResponse)
+	fullTextResponse.Model = modelName
+	usage := model.Usage{
+		PromptTokens:     cohereResponse.Meta.Tokens.InputTokens,
+		CompletionTokens: cohereResponse.Meta.Tokens.OutputTokens,
+		TotalTokens:      cohereResponse.Meta.Tokens.InputTokens + cohereResponse.Meta.Tokens.OutputTokens,
+	}
+	fullTextResponse.Usage = usage
+	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)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &usage
+}

relay/adaptor/cohere/model.go

@@ -0,0 +1,147 @@
+package cohere
+
+type Request struct {
+	Message          string        `json:"message" required:"true"`
+	Model            string        `json:"model,omitempty"`  // 默认值为"command-r"
+	Stream           bool          `json:"stream,omitempty"` // 默认值为false
+	Preamble         string        `json:"preamble,omitempty"`
+	ChatHistory      []ChatMessage `json:"chat_history,omitempty"`
+	ConversationID   string        `json:"conversation_id,omitempty"`
+	PromptTruncation string        `json:"prompt_truncation,omitempty"` // 默认值为"AUTO"
+	Connectors       []Connector   `json:"connectors,omitempty"`
+	Documents        []Document    `json:"documents,omitempty"`
+	Temperature      float64       `json:"temperature,omitempty"` // 默认值为0.3
+	MaxTokens        int           `json:"max_tokens,omitempty"`
+	MaxInputTokens   int           `json:"max_input_tokens,omitempty"`
+	K                int           `json:"k,omitempty"` // 默认值为0
+	P                float64       `json:"p,omitempty"` // 默认值为0.75
+	Seed             int           `json:"seed,omitempty"`
+	StopSequences    []string      `json:"stop_sequences,omitempty"`
+	FrequencyPenalty float64       `json:"frequency_penalty,omitempty"` // 默认值为0.0
+	PresencePenalty  float64       `json:"presence_penalty,omitempty"`  // 默认值为0.0
+	Tools            []Tool        `json:"tools,omitempty"`
+	ToolResults      []ToolResult  `json:"tool_results,omitempty"`
+}
+
+type ChatMessage struct {
+	Role    string `json:"role" required:"true"`
+	Message string `json:"message" required:"true"`
+}
+
+type Tool struct {
+	Name                 string                   `json:"name" required:"true"`
+	Description          string                   `json:"description" required:"true"`
+	ParameterDefinitions map[string]ParameterSpec `json:"parameter_definitions"`
+}
+
+type ParameterSpec struct {
+	Description string `json:"description"`
+	Type        string `json:"type" required:"true"`
+	Required    bool   `json:"required"`
+}
+
+type ToolResult struct {
+	Call    ToolCall                 `json:"call"`
+	Outputs []map[string]interface{} `json:"outputs"`
+}
+
+type ToolCall struct {
+	Name       string                 `json:"name" required:"true"`
+	Parameters map[string]interface{} `json:"parameters" required:"true"`
+}
+
+type StreamResponse struct {
+	IsFinished    bool            `json:"is_finished"`
+	EventType     string          `json:"event_type"`
+	GenerationID  string          `json:"generation_id,omitempty"`
+	SearchQueries []*SearchQuery  `json:"search_queries,omitempty"`
+	SearchResults []*SearchResult `json:"search_results,omitempty"`
+	Documents     []*Document     `json:"documents,omitempty"`
+	Text          string          `json:"text,omitempty"`
+	Citations     []*Citation     `json:"citations,omitempty"`
+	Response      *Response       `json:"response,omitempty"`
+	FinishReason  string          `json:"finish_reason,omitempty"`
+}
+
+type SearchQuery struct {
+	Text         string `json:"text"`
+	GenerationID string `json:"generation_id"`
+}
+
+type SearchResult struct {
+	SearchQuery *SearchQuery `json:"search_query"`
+	DocumentIDs []string     `json:"document_ids"`
+	Connector   *Connector   `json:"connector"`
+}
+
+type Connector struct {
+	ID string `json:"id"`
+}
+
+type Document struct {
+	ID        string `json:"id"`
+	Snippet   string `json:"snippet"`
+	Timestamp string `json:"timestamp"`
+	Title     string `json:"title"`
+	URL       string `json:"url"`
+}
+
+type Citation struct {
+	Start       int      `json:"start"`
+	End         int      `json:"end"`
+	Text        string   `json:"text"`
+	DocumentIDs []string `json:"document_ids"`
+}
+
+type Response struct {
+	ResponseID    string          `json:"response_id"`
+	Text          string          `json:"text"`
+	GenerationID  string          `json:"generation_id"`
+	ChatHistory   []*Message      `json:"chat_history"`
+	FinishReason  *string         `json:"finish_reason"`
+	Meta          Meta            `json:"meta"`
+	Citations     []*Citation     `json:"citations"`
+	Documents     []*Document     `json:"documents"`
+	SearchResults []*SearchResult `json:"search_results"`
+	SearchQueries []*SearchQuery  `json:"search_queries"`
+	Message       string          `json:"message"`
+}
+
+type Message struct {
+	Role    string `json:"role"`
+	Message string `json:"message"`
+}
+
+type Version struct {
+	Version string `json:"version"`
+}
+
+type Units struct {
+	InputTokens  int `json:"input_tokens"`
+	OutputTokens int `json:"output_tokens"`
+}
+
+type ChatEntry struct {
+	Role    string `json:"role"`
+	Message string `json:"message"`
+}
+
+type Meta struct {
+	APIVersion  APIVersion  `json:"api_version"`
+	BilledUnits BilledUnits `json:"billed_units"`
+	Tokens      Usage       `json:"tokens"`
+}
+
+type APIVersion struct {
+	Version string `json:"version"`
+}
+
+type BilledUnits struct {
+	InputTokens  int `json:"input_tokens"`
+	OutputTokens int `json:"output_tokens"`
+}
+
+type Usage struct {
+	InputTokens  int `json:"input_tokens"`
+	OutputTokens int `json:"output_tokens"`
+}

relay/adaptor/common.go

@@ -4,7 +4,7 @@ import (
 	"errors"
 	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/songquanpeng/one-api/relay/client"
+	"github.com/songquanpeng/one-api/common/client"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"io"
 	"net/http"

relay/adaptor/coze/adaptor.go

@@ -0,0 +1,75 @@
+package coze
+
+import (
+	"errors"
+	"fmt"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/relay/adaptor"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+)
+
+type Adaptor struct {
+	meta *meta.Meta
+}
+
+func (a *Adaptor) Init(meta *meta.Meta) {
+	a.meta = meta
+}
+
+func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
+	return fmt.Sprintf("%s/open_api/v2/chat", meta.BaseURL), nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "Bearer "+meta.APIKey)
+	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")
+	}
+	request.User = a.meta.Config.UserID
+	return ConvertRequest(*request), 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 adaptor.DoRequestHelper(a, c, meta, requestBody)
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
+	var responseText *string
+	if meta.IsStream {
+		err, responseText = StreamHandler(c, resp)
+	} else {
+		err, responseText = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+	}
+	if responseText != nil {
+		usage = openai.ResponseText2Usage(*responseText, meta.ActualModelName, meta.PromptTokens)
+	} else {
+		usage = &model.Usage{}
+	}
+	usage.PromptTokens = meta.PromptTokens
+	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "coze"
+}

relay/adaptor/coze/constant/contenttype/define.go

@@ -0,0 +1,5 @@
+package contenttype
+
+const (
+	Text = "text"
+)

relay/adaptor/coze/constant/event/define.go

@@ -0,0 +1,7 @@
+package event
+
+const (
+	Message = "message"
+	Done    = "done"
+	Error   = "error"
+)

relay/adaptor/coze/constant/messagetype/define.go

@@ -0,0 +1,6 @@
+package messagetype
+
+const (
+	Answer   = "answer"
+	FollowUp = "follow_up"
+)

relay/adaptor/coze/constants.go

@@ -0,0 +1,3 @@
+package coze
+
+var ModelList = []string{}

relay/adaptor/coze/helper.go

@@ -0,0 +1,10 @@
+package coze
+
+import "github.com/songquanpeng/one-api/relay/adaptor/coze/constant/event"
+
+func event2StopReason(e *string) string {
+	if e == nil || *e == event.Message {
+		return ""
+	}
+	return "stop"
+}

relay/adaptor/coze/main.go

@@ -0,0 +1,202 @@
+package coze
+
+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"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/logger"
+	"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"
+)
+
+// https://www.coze.com/open
+
+func stopReasonCoze2OpenAI(reason *string) string {
+	if reason == nil {
+		return ""
+	}
+	switch *reason {
+	case "end_turn":
+		return "stop"
+	case "stop_sequence":
+		return "stop"
+	case "max_tokens":
+		return "length"
+	default:
+		return *reason
+	}
+}
+
+func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request {
+	cozeRequest := Request{
+		Stream: textRequest.Stream,
+		User:   textRequest.User,
+		BotId:  strings.TrimPrefix(textRequest.Model, "bot-"),
+	}
+	for i, message := range textRequest.Messages {
+		if i == len(textRequest.Messages)-1 {
+			cozeRequest.Query = message.StringContent()
+			continue
+		}
+		cozeMessage := Message{
+			Role:    message.Role,
+			Content: message.StringContent(),
+		}
+		cozeRequest.ChatHistory = append(cozeRequest.ChatHistory, cozeMessage)
+	}
+	return &cozeRequest
+}
+
+func StreamResponseCoze2OpenAI(cozeResponse *StreamResponse) (*openai.ChatCompletionsStreamResponse, *Response) {
+	var response *Response
+	var stopReason string
+	var choice openai.ChatCompletionsStreamResponseChoice
+
+	if cozeResponse.Message != nil {
+		if cozeResponse.Message.Type != messagetype.Answer {
+			return nil, nil
+		}
+		choice.Delta.Content = cozeResponse.Message.Content
+	}
+	choice.Delta.Role = "assistant"
+	finishReason := stopReasonCoze2OpenAI(&stopReason)
+	if finishReason != "null" {
+		choice.FinishReason = &finishReason
+	}
+	var openaiResponse openai.ChatCompletionsStreamResponse
+	openaiResponse.Object = "chat.completion.chunk"
+	openaiResponse.Choices = []openai.ChatCompletionsStreamResponseChoice{choice}
+	openaiResponse.Id = cozeResponse.ConversationId
+	return &openaiResponse, response
+}
+
+func ResponseCoze2OpenAI(cozeResponse *Response) *openai.TextResponse {
+	var responseText string
+	for _, message := range cozeResponse.Messages {
+		if message.Type == messagetype.Answer {
+			responseText = message.Content
+			break
+		}
+	}
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    "assistant",
+			Content: responseText,
+			Name:    nil,
+		},
+		FinishReason: "stop",
+	}
+	fullTextResponse := openai.TextResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", cozeResponse.ConversationId),
+		Model:   "coze-bot",
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+	}
+	return &fullTextResponse
+}
+
+func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *string) {
+	var responseText string
+	createdTime := helper.GetTimestamp()
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(bufio.ScanLines)
+
+	common.SetEventStreamHeaders(c)
+	var modelName string
+
+	for scanner.Scan() {
+		data := scanner.Text()
+		if len(data) < 5 || !strings.HasPrefix(data, "data:") {
+			continue
+		}
+		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
+	}
+
+	return nil, &responseText
+}
+
+func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *string) {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	var cozeResponse Response
+	err = json.Unmarshal(responseBody, &cozeResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	if cozeResponse.Code != 0 {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: cozeResponse.Msg,
+				Code:    cozeResponse.Code,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	fullTextResponse := ResponseCoze2OpenAI(&cozeResponse)
+	fullTextResponse.Model = modelName
+	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)
+	_, err = c.Writer.Write(jsonResponse)
+	var responseText string
+	if len(fullTextResponse.Choices) > 0 {
+		responseText = fullTextResponse.Choices[0].Message.StringContent()
+	}
+	return nil, &responseText
+}

relay/adaptor/coze/model.go

@@ -0,0 +1,38 @@
+package coze
+
+type Message struct {
+	Role        string `json:"role"`
+	Type        string `json:"type"`
+	Content     string `json:"content"`
+	ContentType string `json:"content_type"`
+}
+
+type ErrorInformation struct {
+	Code int    `json:"code"`
+	Msg  string `json:"msg"`
+}
+
+type Request struct {
+	ConversationId string    `json:"conversation_id,omitempty"`
+	BotId          string    `json:"bot_id"`
+	User           string    `json:"user"`
+	Query          string    `json:"query"`
+	ChatHistory    []Message `json:"chat_history,omitempty"`
+	Stream         bool      `json:"stream"`
+}
+
+type Response struct {
+	ConversationId string    `json:"conversation_id,omitempty"`
+	Messages       []Message `json:"messages,omitempty"`
+	Code           int       `json:"code,omitempty"`
+	Msg            string    `json:"msg,omitempty"`
+}
+
+type StreamResponse struct {
+	Event            string            `json:"event,omitempty"`
+	Message          *Message          `json:"message,omitempty"`
+	IsFinish         bool              `json:"is_finish,omitempty"`
+	Index            int               `json:"index,omitempty"`
+	ConversationId   string            `json:"conversation_id,omitempty"`
+	ErrorInformation *ErrorInformation `json:"error_information,omitempty"`
+}

relay/adaptor/deepl/adaptor.go

@@ -1,32 +1,32 @@
-package aws
+package deepl
 
 import (
-	"github.com/songquanpeng/one-api/common/ctxkey"
-	"io"
-	"net/http"
-
+	"errors"
+	"fmt"
 	"github.com/gin-gonic/gin"
-	"github.com/pkg/errors"
 	"github.com/songquanpeng/one-api/relay/adaptor"
-	"github.com/songquanpeng/one-api/relay/adaptor/anthropic"
 	"github.com/songquanpeng/one-api/relay/meta"
 	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
 )
 
-var _ adaptor.Adaptor = new(Adaptor)
-
 type Adaptor struct {
+	meta       *meta.Meta
+	promptText string
 }
 
 func (a *Adaptor) Init(meta *meta.Meta) {
-
+	a.meta = meta
 }
 
 func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
-	return "", nil
+	return fmt.Sprintf("%s/v2/translate", meta.BaseURL), nil
 }
 
 func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", "DeepL-Auth-Key "+meta.APIKey)
 	return nil
 }
 
@@ -34,11 +34,9 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	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
+	convertedRequest, text := ConvertRequest(*request)
+	a.promptText = text
+	return convertedRequest, nil
 }
 
 func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
@@ -49,26 +47,27 @@ 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
+	return adaptor.DoRequestHelper(a, c, meta, requestBody)
 }
 
 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, resp)
+		err = StreamHandler(c, resp, meta.ActualModelName)
 	} else {
-		err, usage = Handler(c, resp, meta.PromptTokens, meta.ActualModelName)
+		err = Handler(c, resp, meta.ActualModelName)
+	}
+	promptTokens := len(a.promptText)
+	usage = &model.Usage{
+		PromptTokens: promptTokens,
+		TotalTokens:  promptTokens,
 	}
 	return
 }
 
-func (a *Adaptor) GetModelList() (models []string) {
-	for n := range awsModelIDMap {
-		models = append(models, n)
-	}
-
-	return
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
 }
 
 func (a *Adaptor) GetChannelName() string {
-	return "aws"
+	return "deepl"
 }

relay/adaptor/deepl/constants.go

@@ -0,0 +1,9 @@
+package deepl
+
+// https://developers.deepl.com/docs/api-reference/glossaries
+
+var ModelList = []string{
+	"deepl-zh",
+	"deepl-en",
+	"deepl-ja",
+}

relay/adaptor/deepl/helper.go

@@ -0,0 +1,11 @@
+package deepl
+
+import "strings"
+
+func parseLangFromModelName(modelName string) string {
+	parts := strings.Split(modelName, "-")
+	if len(parts) == 1 {
+		return "ZH"
+	}
+	return parts[1]
+}

relay/adaptor/deepl/main.go

@@ -0,0 +1,137 @@
+package deepl
+
+import (
+	"encoding/json"
+	"github.com/gin-gonic/gin"
+	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/constant/finishreason"
+	"github.com/songquanpeng/one-api/relay/constant/role"
+	"github.com/songquanpeng/one-api/relay/model"
+	"io"
+	"net/http"
+)
+
+// https://developers.deepl.com/docs/getting-started/your-first-api-request
+
+func ConvertRequest(textRequest model.GeneralOpenAIRequest) (*Request, string) {
+	var text string
+	if len(textRequest.Messages) != 0 {
+		text = textRequest.Messages[len(textRequest.Messages)-1].StringContent()
+	}
+	deeplRequest := Request{
+		TargetLang: parseLangFromModelName(textRequest.Model),
+		Text:       []string{text},
+	}
+	return &deeplRequest, text
+}
+
+func StreamResponseDeepL2OpenAI(deeplResponse *Response) *openai.ChatCompletionsStreamResponse {
+	var choice openai.ChatCompletionsStreamResponseChoice
+	if len(deeplResponse.Translations) != 0 {
+		choice.Delta.Content = deeplResponse.Translations[0].Text
+	}
+	choice.Delta.Role = role.Assistant
+	choice.FinishReason = &constant.StopFinishReason
+	openaiResponse := openai.ChatCompletionsStreamResponse{
+		Object:  constant.StreamObject,
+		Created: helper.GetTimestamp(),
+		Choices: []openai.ChatCompletionsStreamResponseChoice{choice},
+	}
+	return &openaiResponse
+}
+
+func ResponseDeepL2OpenAI(deeplResponse *Response) *openai.TextResponse {
+	var responseText string
+	if len(deeplResponse.Translations) != 0 {
+		responseText = deeplResponse.Translations[0].Text
+	}
+	choice := openai.TextResponseChoice{
+		Index: 0,
+		Message: model.Message{
+			Role:    role.Assistant,
+			Content: responseText,
+			Name:    nil,
+		},
+		FinishReason: finishreason.Stop,
+	}
+	fullTextResponse := openai.TextResponse{
+		Object:  constant.NonStreamObject,
+		Created: helper.GetTimestamp(),
+		Choices: []openai.TextResponseChoice{choice},
+	}
+	return &fullTextResponse
+}
+
+func StreamHandler(c *gin.Context, resp *http.Response, modelName string) *model.ErrorWithStatusCode {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError)
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError)
+	}
+	var deeplResponse Response
+	err = json.Unmarshal(responseBody, &deeplResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError)
+	}
+	fullTextResponse := StreamResponseDeepL2OpenAI(&deeplResponse)
+	fullTextResponse.Model = modelName
+	fullTextResponse.Id = helper.GetResponseID(c)
+	jsonData, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError)
+	}
+	common.SetEventStreamHeaders(c)
+	c.Stream(func(w io.Writer) bool {
+		if jsonData != nil {
+			c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonData)})
+			jsonData = nil
+			return true
+		}
+		c.Render(-1, common.CustomEvent{Data: "data: [DONE]"})
+		return false
+	})
+	_ = resp.Body.Close()
+	return nil
+}
+
+func Handler(c *gin.Context, resp *http.Response, modelName string) *model.ErrorWithStatusCode {
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError)
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError)
+	}
+	var deeplResponse Response
+	err = json.Unmarshal(responseBody, &deeplResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError)
+	}
+	if deeplResponse.Message != "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: deeplResponse.Message,
+				Code:    "deepl_error",
+			},
+			StatusCode: resp.StatusCode,
+		}
+	}
+	fullTextResponse := ResponseDeepL2OpenAI(&deeplResponse)
+	fullTextResponse.Model = modelName
+	fullTextResponse.Id = helper.GetResponseID(c)
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError)
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil
+}

relay/adaptor/deepl/model.go

@@ -0,0 +1,16 @@
+package deepl
+
+type Request struct {
+	Text       []string `json:"text"`
+	TargetLang string   `json:"target_lang"`
+}
+
+type Translation struct {
+	DetectedSourceLanguage string `json:"detected_source_language,omitempty"`
+	Text                   string `json:"text,omitempty"`
+}
+
+type Response struct {
+	Translations []Translation `json:"translations,omitempty"`
+	Message      string        `json:"message,omitempty"`
+}

relay/adaptor/deepseek/constants.go

@@ -0,0 +1,6 @@
+package deepseek
+
+var ModelList = []string{
+	"deepseek-chat",
+	"deepseek-coder",
+}

relay/adaptor/doubao/constants.go

@@ -0,0 +1,13 @@
+package doubao
+
+// https://console.volcengine.com/ark/region:ark+cn-beijing/model
+
+var ModelList = []string{
+	"Doubao-pro-128k",
+	"Doubao-pro-32k",
+	"Doubao-pro-4k",
+	"Doubao-lite-128k",
+	"Doubao-lite-32k",
+	"Doubao-lite-4k",
+	"Doubao-embedding",
+}