biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[ROCm] Add aiter tkw1 kernel for Llama4 fp8 (#16727)

Browse Source 
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: vllmellm <vllm.ellm@embeddedllm.com>
This commit is contained in:
kliuae
2025-04-22 11:42:34 +08:00
committed by GitHub
parent 0e4254492f
commit 5b794cae8d
6 changed files with 134 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docker/Dockerfile.rocm_base
View File

@@ -12,7 +12,7 @@ ARG PYTORCH_REPO="https://github.com/pytorch/pytorch.git"
ARG PYTORCH_VISION_REPO="https://github.com/pytorch/vision.git"
ARG FA_BRANCH="1a7f4dfa"
ARG FA_REPO="https://github.com/Dao-AILab/flash-attention.git"
ARG AITER_BRANCH="8970b25b"
ARG AITER_BRANCH="5a77249"
ARG AITER_REPO="https://github.com/ROCm/aiter.git"
FROM ${BASE_IMAGE} AS base