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 15, 12:15 UTC
[Bugfix][ROCm] Restrict ray version due to a breaking release (#17480)
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Gregory Shtrasberg committed
Apr 30, 2025 at 12:59 UTC
584f5fb4c6d96365a3bfa8594115bc02744f2096
1 file changed
+1
-1
requirements/rocm.txt
+1
-1
@@ -9,7 +9,7 @@ awscli
9
boto3
10
botocore
11
datasets
12
-ray >= 2.10.0
12
+ray>=2.10.0,<2.45.0
13
peft
14
pytest-asyncio
15
tensorizer>=2.9.0