A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[ROCm] Update Triton, Torch, and AITER branches for ROCm base Dockerfile (#27206)
Signed-off-by: Micah Williamson <micah.williamson@amd.com>
Micah Williamson committed
Oct 21, 2025 at 11:01 UTC
aa1356ec53a65a79a0027e8c265c76d84de8c046
1 file changed
+3
-3
docker/Dockerfile.rocm_base
+3
-3
@@ -1,13 +1,13 @@
1
ARG BASE_IMAGE=rocm/dev-ubuntu-22.04:7.0-complete
2
-ARG TRITON_BRANCH="f9e5bf54"
2
+ARG TRITON_BRANCH="57c693b6"
3
ARG TRITON_REPO="https://github.com/ROCm/triton.git"
4
-ARG PYTORCH_BRANCH="b2fb6885"
4
+ARG PYTORCH_BRANCH="1c57644d"
5
ARG PYTORCH_VISION_BRANCH="v0.23.0"
6
ARG PYTORCH_REPO="https://github.com/ROCm/pytorch.git"
7
ARG PYTORCH_VISION_REPO="https://github.com/pytorch/vision.git"
8
ARG FA_BRANCH="0e60e394"
9
ARG FA_REPO="https://github.com/Dao-AILab/flash-attention.git"
10
-ARG AITER_BRANCH="2ab9f4cd"
10
+ARG AITER_BRANCH="eef23c7f"
11
ARG AITER_REPO="https://github.com/ROCm/aiter.git"
12
13
FROM ${BASE_IMAGE} AS base