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 13, 08:15 UTC
[ROCm][Bugfix] Revert removing setuptools version restriction (#28592)
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Gregory Shtrasberg committed
Nov 12, 2025 at 19:46 UTC
d75ad048184f2864dad6b5da2e6e11322aec13aa
2 files changed
+2
-2
requirements/rocm-build.txt
+1
-1
@@ -9,7 +9,7 @@ torchaudio==2.9.0
9
triton==3.5.0
10
cmake>=3.26.1,<4
11
packaging>=24.2
12
-setuptools>=77.0.3,<81.0.0
12
+setuptools>=77.0.3,<80.0.0
13
setuptools-scm>=8
14
wheel
15
jinja2>=3.1.6
requirements/rocm.txt
+1
-1
@@ -10,7 +10,7 @@ peft
10
pytest-asyncio
11
tensorizer==2.10.1
12
packaging>=24.2
13
-setuptools>=77.0.3,<81.0.0
13
+setuptools>=77.0.3,<80.0.0
14
setuptools-scm>=8
15
runai-model-streamer[s3,gcs]==0.15.0
16
conch-triton-kernels==1.2.1