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 16, 23:15 UTC
[BugFix] exclude version 1.15.0 for modelscope (#5668)
zhyncs committed
Jun 22, 2024 at 03:15 UTC
f1e72cc19a21928400b63743d5fe164ec8ed30e8
1 file changed
+1
-1
Dockerfile
+1
-1
@@ -172,7 +172,7 @@ FROM vllm-base AS vllm-openai
172
173
# install additional dependencies for openai api server
174
RUN --mount=type=cache,target=/root/.cache/pip \
175
- pip install accelerate hf_transfer modelscope
175
+ pip install accelerate hf_transfer 'modelscope!=1.15.0'
176
177
ENV VLLM_USAGE_SOURCE production-docker-image
178