Merge branch 'main' into cleanup-req

Dockerfile

@@ -128,22 +128,9 @@ FROM vllm-base AS test
 
 ADD . /vllm-workspace/
 
-
-#################### RUNTIME BASE IMAGE ####################
-# We used base cuda image because pytorch installs its own cuda libraries.
-# However pynccl depends on cuda libraries so we had to switch to the runtime image
-# In the future it would be nice to get a container with pytorch and cuda without duplicating cuda
-FROM nvidia/cuda:12.1.0-runtime-ubuntu22.04 AS vllm-base
-
-# libnccl required for ray
-RUN apt-get update -y \
-    && apt-get install -y python3-pip
-
-WORKDIR /workspace
-COPY requirements-common.txt requirements-common.txt
-COPY requirements-cuda.txt requirements-cuda.txt
+# install development dependencies (for testing)
 RUN --mount=type=cache,target=/root/.cache/pip \
-    pip install -r requirements-cuda.txt
+    pip install -r requirements-dev.txt
 
 # doc requires source code
 # we hide them inside `test_docs/` , so that this source code