text-generation-inference/server
drbh bdc47394d2 feat: support phi3.5 moe (#2479)
* feat: support phi3.5 moe model loading

* fix: prefer llama base model and improve rotary logic

* feat: return reasonable generation and add integration test

* fix: run lint and update docs

* fix: rerun lint for openapi docs

* fix: prefer do_sample false unless temp is set by user, and update chat tests

* fix: small typo adjustments

* fix: consolidate long rope paths

* fix: revert greedy by default and test changes

* Vendor configuration so that we don't have to `trust_remote_code`

* Use SparseMoELayer

* Add support for dense MoE

* Some type annotations

* Add the usual model tests

* Ruff.

---------

Co-authored-by: Daniël de Kok <me@danieldk.eu>
Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>
2024-10-25 09:12:03 +00:00
..
custom_kernels All integration tests back everywhere (too many failed CI). (#2428) 2024-09-25 06:10:59 +00:00
exllama_kernels Update ROCM libs and improvements (#2579) 2024-10-25 09:01:04 +00:00
exllamav2_kernels Update ROCM libs and improvements (#2579) 2024-10-25 09:01:04 +00:00
tests Fix tokenization yi (#2507) 2024-09-25 06:15:35 +00:00
text_generation_server feat: support phi3.5 moe (#2479) 2024-10-25 09:12:03 +00:00
.gitignore Impl simple mamba model (#1480) 2024-04-23 11:45:11 +03:00
dill-0.3.7-patch.sh Make Gaudi adapt to the tgi 2.3.0 2024-09-26 06:04:55 +00:00
dill-0.3.8-patch.sh Make Gaudi adapt to the tgi 2.3.0 2024-09-26 06:04:55 +00:00
Makefile Make Gaudi adapt to the tgi 2.3.0 2024-09-26 06:04:55 +00:00
Makefile-awq chore: add pre-commit (#1569) 2024-04-24 15:32:02 +03:00
Makefile-eetq Upgrade EETQ (Fixes the cuda graphs). (#1729) 2024-04-25 17:58:27 +03:00
Makefile-exllamav2 Upgrading exl2. (#2415) 2024-09-25 06:07:40 +00:00
Makefile-fbgemm Add Directory Check to Prevent Redundant Cloning in Build Process (#2486) 2024-09-25 06:14:07 +00:00
Makefile-flash-att Hotfixing make install. (#2008) 2024-09-24 03:29:29 +00:00
Makefile-flash-att-v2 Update ROCM libs and improvements (#2579) 2024-10-25 09:01:04 +00:00
Makefile-flashinfer Prefix test - Different kind of load test to trigger prefix test bugs. (#2490) 2024-09-25 06:14:07 +00:00
Makefile-lorax-punica Enable multiple LoRa adapters (#2010) 2024-09-24 03:55:04 +00:00
Makefile-selective-scan chore: add pre-commit (#1569) 2024-04-24 15:32:02 +03:00
Makefile-vllm Update ROCM libs and improvements (#2579) 2024-10-25 09:01:04 +00:00
poetry.lock Add support for GPTQ-quantized MoE models using MoE Marlin (#2557) 2024-10-25 09:07:52 +00:00
pyproject.toml Make Gaudi adapt to the tgi 2.3.0 2024-09-26 06:04:55 +00:00
README.md chore: add pre-commit (#1569) 2024-04-24 15:32:02 +03:00
requirements_cuda.txt hotfix: add syrupy to the right subproject (#2499) 2024-09-25 06:13:36 +00:00
requirements_intel.txt hotfix: add syrupy to the right subproject (#2499) 2024-09-25 06:13:36 +00:00
requirements_rocm.txt hotfix: add syrupy to the right subproject (#2499) 2024-09-25 06:13:36 +00:00
requirements.txt Make Gaudi adapt to the tgi 2.3.0 2024-09-26 06:04:55 +00:00

Text Generation Inference Python gRPC Server

A Python gRPC server for Text Generation Inference

Install

make install

Run

make run-dev