owen/worker-vllm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat: update to 0.23.0

Browse Source
This commit is contained in:
velaraptor-runpod
2026-07-09 18:22:40 -05:00
parent 798f2b12eb
commit 88490e9551
2 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Dockerfile
+2 -1
View File
@@ -16,7 +16,8 @@ ENV PATH="/root/.local/bin:$PATH"
RUN ldconfig /usr/local/cuda-13.0/compat/ RUN ldconfig /usr/local/cuda-13.0/compat/
# Install vLLM with FlashInfer - use CUDA 130 PyTorch wheels # Install vLLM with FlashInfer - use CUDA 130 PyTorch wheels
RUN uv pip install --system "packaging>=24.2" && \ RUN --mount=type=cache,target=/root/.cache/uv \
uv pip install --system "packaging>=24.2" && \
uv pip install --system "vllm[flashinfer]==0.23.0" && \ uv pip install --system "vllm[flashinfer]==0.23.0" && \
uv pip install --system git+https://github.com/deepseek-ai/DeepGEMM.git@714dd1a4a980f7937a74343d19a8eba4fe321480 --no-build-isolation && \ uv pip install --system git+https://github.com/deepseek-ai/DeepGEMM.git@714dd1a4a980f7937a74343d19a8eba4fe321480 --no-build-isolation && \
uv pip install --system --force-reinstall --no-deps nixl-cu13 uv pip install --system --force-reinstall --no-deps nixl-cu13
configs/qwen/qwen3_8b.yaml
+2
View File
@@ -7,4 +7,6 @@ quantization: fp8
kv-cache-dtype: fp8 kv-cache-dtype: fp8
enforce-eager: false enforce-eager: false
enable-prefix-caching: true enable-prefix-caching: true
vllm-release: v2.22.5
compilation-config: '{"cudagraph_mode": "PIECEWISE"}'
speculative-config: '{"model":"RedHatAI/Qwen3-8B-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}' speculative-config: '{"model":"RedHatAI/Qwen3-8B-speculator.eagle3","method":"eagle3","num_speculative_tokens":3}'