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 11, 01:15 UTC
[ROCm][CI/Build] Fix triton version to one that has triton_kernels required for gpt-oss to run (#31159)
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Gregory Shtrasberg committed
Dec 22, 2025 at 11:19 UTC
ab3a85fd688dc2544edab3098fab8e2c8893a707
1 file changed
+2
-2
docker/Dockerfile.rocm_base
+2
-2
@@ -1,5 +1,5 @@
1
ARG BASE_IMAGE=rocm/dev-ubuntu-22.04:7.0-complete
2
-ARG TRITON_BRANCH="a272dfa8"
2
+ARG TRITON_BRANCH="57c693b6"
3
ARG TRITON_REPO="https://github.com/ROCm/triton.git"
4
ARG PYTORCH_BRANCH="89075173"
5
ARG PYTORCH_REPO="https://github.com/ROCm/pytorch.git"
@@ -162,4 +162,4 @@ RUN echo "BASE_IMAGE: ${BASE_IMAGE}" > /app/versions.txt \
162
&& echo "FA_BRANCH: ${FA_BRANCH}" >> /app/versions.txt \
163
&& echo "FA_REPO: ${FA_REPO}" >> /app/versions.txt \
164
&& echo "AITER_BRANCH: ${AITER_BRANCH}" >> /app/versions.txt \
165
- && echo "AITER_REPO: ${AITER_REPO}" >> /app/versions.txt
\ No newline at end of file
165
+ && echo "AITER_REPO: ${AITER_REPO}" >> /app/versions.txt