feat: bump vLLM version to v0.8.4 (#690)

container/Dockerfile.vllm

@@ -181,16 +181,12 @@ RUN mkdir /opt/dynamo && \
 ENV VIRTUAL_ENV=/opt/dynamo/venv
 ENV PATH="${VIRTUAL_ENV}/bin:${PATH}"
 
-# Common dependencies
-RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
-    uv pip install --requirement /tmp/requirements.txt
-
 # Install patched vllm - keep this early in Dockerfile to avoid
 # rebuilds from unrelated source code changes
-ARG VLLM_REF="0.7.2"
+ARG VLLM_REF="0.8.4"
 ARG VLLM_PATCH="vllm_v${VLLM_REF}-dynamo-kv-disagg-patch.patch"
 ARG VLLM_PATCHED_PACKAGE_NAME="ai_dynamo_vllm"
-ARG VLLM_PATCHED_PACKAGE_VERSION="0.7.2.post1"
+ARG VLLM_PATCHED_PACKAGE_VERSION="0.8.4"
 RUN --mount=type=bind,source=./container/deps/,target=/tmp/deps \
     mkdir /tmp/vllm && \
     uv pip install pip wheel && \
@@ -211,6 +207,10 @@ RUN --mount=type=bind,source=./container/deps/,target=/tmp/deps \
     wheel pack . --dest-dir /workspace/dist && \
     uv pip install /workspace/dist/${VLLM_PATCHED_PACKAGE_NAME}-*.whl
 
+# Common dependencies
+RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
+    uv pip install --requirement /tmp/requirements.txt
+
 # Install test dependencies
 RUN --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.txt \
     uv pip install --requirement /tmp/requirements.txt

container/deps/vllm/tests/test_patch_install.py

@@ -27,4 +27,4 @@ pytestmark = pytest.mark.pre_merge
 @pytest.mark.skipif(vllm is None, reason="Skipping vllm tests, vllm not installed")
 def test_version():
     # Verify that the image has the patched version of vllm
-    assert vllm.__version__.endswith("0.7.2")  # type: ignore
+    assert vllm.__version__.endswith("0.8.4")  # type: ignore