FROM rocm/vllm:rocm7.12.0_gfx120X-all_ubuntu24.04_py3.12_pytorch_2.9.1_vllm_0.16.0 WORKDIR /workspace COPY requirements.txt /workspace/requirements.txt RUN pip install --no-cache-dir -r /workspace/requirements.txt COPY app /workspace/app ENV PYTHONUNBUFFERED=1 ENV HOST=0.0.0.0 ENV PORT=8000 ENV TENSOR_PARALLEL_SIZE=2 ENV GPU_MEMORY_UTILIZATION=0.92 ENV MAX_MODEL_LEN=8192 ENV MAX_NUM_SEQS=64 ENV DTYPE=bfloat16 ENV ENFORCE_EAGER=false ENV TRUST_REMOTE_CODE=false EXPOSE 8000 8001 CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]