chore: vllm 0.10.1.1 (#2641)

Signed-off-by: Jason Zhou <[email protected]>

Author: dmitry-tokarev-nv

container/Dockerfile.vllm

@@ -13,11 +13,11 @@ ARG RUNTIME_IMAGE="nvcr.io/nvidia/cuda"
 ARG RUNTIME_IMAGE_TAG="12.8.1-runtime-ubuntu24.04"
 
 # Make sure to update the dependency version in pyproject.toml when updating this
-ARG VLLM_REF="aab549870df50edf0512f0a59b574f692f546465"  # from v0.10.1
+ARG VLLM_REF="1da94e673c257373280026f75ceb4effac80e892"  # from v0.10.1.1
 ARG TORCH_BACKEND="cu128"
 
-# Match 0.10.1 vLLM release
-# https://github.com/vllm-project/vllm/releases/tag/v0.10.1
+# Match 0.10.1.1 vLLM release
+# https://github.com/vllm-project/vllm/releases/tag/v0.10.1.1
 # Pinned to commit before https://github.com/deepseek-ai/DeepGEMM/pull/112 for DeepGEMM which seems to break on H100:
 # "RuntimeError: Failed: CUDA runtime error csrc/jit/kernel_runtime.hpp:108 '98'"
 ARG DEEPGEMM_REF="f85ec64"

container/deps/vllm/install_vllm.sh

@@ -20,10 +20,10 @@ set -euo pipefail
 
 # Parse arguments
 EDITABLE=true
-VLLM_REF="aab549870df50edf0512f0a59b574f692f546465"  # from v0.10.1
+VLLM_REF="1da94e673c257373280026f75ceb4effac80e892"  # from v0.10.1.1
 # When updating above VLLM_REF make sure precompiled wheel file URL is correct. Run this command:
 # aws s3 ls s3://vllm-wheels/${VLLM_REF}/ --region us-west-2 --no-sign-request
-VLLM_PRECOMPILED_WHEEL_LOCATION="https://vllm-wheels.s3.us-west-2.amazonaws.com/${VLLM_REF}/vllm-0.10.1-cp38-abi3-manylinux1_x86_64.whl"
+VLLM_PRECOMPILED_WHEEL_LOCATION="https://vllm-wheels.s3.us-west-2.amazonaws.com/${VLLM_REF}/vllm-0.10.1.1-cp38-abi3-manylinux1_x86_64.whl"
 VLLM_GIT_URL="https://github.com/vllm-project/vllm.git"
 MAX_JOBS=16
 INSTALLATION_DIR=/tmp
@@ -86,13 +86,13 @@ while [[ $# -gt 0 ]]; do
             echo "Options:"
             echo "  --editable        Install vllm in editable mode (default)"
             echo "  --no-editable     Install vllm in non-editable mode"
-            echo f"  --vllm-ref REF    Git reference to checkout (default: ${VLLM_REF})"
-            echo f"  --max-jobs NUM    Maximum number of parallel jobs (default: ${MAX_JOBS})"
+            echo "  --vllm-ref REF    Git reference to checkout (default: ${VLLM_REF})"
+            echo "  --max-jobs NUM    Maximum number of parallel jobs (default: ${MAX_JOBS})"
             echo "  --arch ARCH       Architecture (amd64|arm64, default: auto-detect)"
-            echo f"  --installation-dir DIR  Directory to install vllm (default: ${INSTALLATION_DIR})"
-            echo f"  --deepgemm-ref REF  Git reference for DeepGEMM (default: ${DEEPGEMM_REF})"
-            echo f"  --flashinf-ref REF  Git reference for Flash Infer (default: ${FLASHINF_REF})"
-            echo f"  --torch-backend BACKEND  Torch backend to use (default: ${TORCH_BACKEND})"
+            echo "  --installation-dir DIR  Directory to install vllm (default: ${INSTALLATION_DIR})"
+            echo "  --deepgemm-ref REF  Git reference for DeepGEMM (default: ${DEEPGEMM_REF})"
+            echo "  --flashinf-ref REF  Git reference for Flash Infer (default: ${FLASHINF_REF})"
+            echo "  --torch-backend BACKEND  Torch backend to use (default: ${TORCH_BACKEND})"
             exit 0
             ;;
         *)

pyproject.toml

@@ -55,7 +55,7 @@ trtllm =[
 vllm = [
     "uvloop",
     "nixl<=0.4.1",
-    "vllm[flashinfer]==0.10.1",
+    "vllm[flashinfer]==0.10.1.1",
 ]
 
 sglang = [