feat: upgrade vllm to 0.21.0

2026-06-03 14:43:56 -05:00
parent 80072047ab
commit cb3f077dba
2 changed files with 2 additions and 2 deletions
@@ -10,7 +10,7 @@ RUN ldconfig /usr/local/cuda-13.0/compat/

 # Install vLLM with FlashInfer - use CUDA 130 PyTorch wheels
 RUN uv pip install --system "packaging>=24.2" && \
-    uv pip install --system "vllm[flashinfer]==0.20.2" && \
+    uv pip install --system "vllm[flashinfer]==0.21.0" && \
    uv pip install --system git+https://github.com/deepseek-ai/DeepGEMM.git@714dd1a4a980f7937a74343d19a8eba4fe321480 --no-build-isolation

 # Install additional Python dependencies (after vLLM to avoid PyTorch version conflicts)