huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

use ROCm instead of RoCm

Browse Source
This commit is contained in:
Félix Marty 2023-12-04 10:31:01 +01:00
parent bda189062d
commit 529d7d3676
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/supported_models.md
View File

@ -41,7 +41,7 @@ text-generation-launcher --model-id <PATH-TO-LOCAL-BLOOM>
TGI optimized models are supported on NVIDIA [A100](https://www.nvidia.com/en-us/data-center/a100/), [A10G](https://www.nvidia.com/en-us/data-center/products/a10-gpu/) and [T4](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPUs with CUDA 11.8+. Note that you have to install [NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html) to use it. For other NVIDIA GPUs, continuous batching will still apply, but some operations like flash attention and paged attention will not be executed. TGI optimized models are supported on NVIDIA [A100](https://www.nvidia.com/en-us/data-center/a100/), [A10G](https://www.nvidia.com/en-us/data-center/products/a10-gpu/) and [T4](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPUs with CUDA 11.8+. Note that you have to install [NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html) to use it. For other NVIDIA GPUs, continuous batching will still apply, but some operations like flash attention and paged attention will not be executed.
TGI also has support of ROCm-enabled AMD Instinct MI210 and MI250 GPUs, with paged attention and flash attention v2 support. The following features are currently not supported in the RoCm version of TGI, and the supported may be extended in the future: TGI also has support of ROCm-enabled AMD Instinct MI210 and MI250 GPUs, with paged attention and flash attention v2 support. The following features are currently not supported in the ROCm version of TGI, and the supported may be extended in the future:
* quantization (GPTQ, AWQ, etc.) * quantization (GPTQ, AWQ, etc.)
* Flash [layer norm kernel](https://github.com/Dao-AILab/flash-attention/tree/main/csrc/layer_norm) * Flash [layer norm kernel](https://github.com/Dao-AILab/flash-attention/tree/main/csrc/layer_norm)
* Kernel for slinding window attention (Mistral) * Kernel for slinding window attention (Mistral)