+3
-24
@@ -580,33 +580,12 @@ RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
# Install FlashInfer JIT cache (requires CUDA-version-specific index URL)
|
||||
# https://docs.flashinfer.ai/installation.html
|
||||
# From versions.json: .flashinfer.version
|
||||
# 0.6.7: CUTLASS 4.4.2 bump, fixes TMA grouped GEMM on SM12x (flashinfer#2798)
|
||||
# TODO: bump to 0.6.8 when released for NVFP4/MXFP4 group GEMMs on
|
||||
# SM120/SM121 (RTX 50 / DGX Spark) via flashinfer#2738
|
||||
ARG FLASHINFER_VERSION=0.6.7
|
||||
ARG FLASHINFER_VERSION=0.6.8.post1
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} \
|
||||
--extra-index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') \
|
||||
&& flashinfer show-config
|
||||
|
||||
# Pre-download FlashInfer TRTLLM BMM headers for air-gapped environments.
|
||||
# At runtime, MoE JIT compilation downloads these from edge.urm.nvidia.com
|
||||
# which fails without internet. This step caches them at build time.
|
||||
RUN python3 <<'PYEOF'
|
||||
from flashinfer.jit import env as jit_env
|
||||
from flashinfer.jit.cubin_loader import download_trtllm_headers, get_cubin
|
||||
from flashinfer.artifacts import ArtifactPath, CheckSumHash
|
||||
|
||||
download_trtllm_headers(
|
||||
'bmm',
|
||||
jit_env.FLASHINFER_CUBIN_DIR / 'flashinfer' / 'trtllm' / 'batched_gemm' / 'trtllmGen_bmm_export',
|
||||
f'{ArtifactPath.TRTLLM_GEN_BMM}/include/trtllmGen_bmm_export',
|
||||
ArtifactPath.TRTLLM_GEN_BMM,
|
||||
get_cubin(f'{ArtifactPath.TRTLLM_GEN_BMM}/checksums.txt', CheckSumHash.TRTLLM_GEN_BMM),
|
||||
)
|
||||
|
||||
print('FlashInfer TRTLLM BMM headers downloaded successfully')
|
||||
PYEOF
|
||||
&& flashinfer show-config \
|
||||
&& flashinfer download-cubin
|
||||
|
||||
# ============================================================
|
||||
# OPENAI API SERVER DEPENDENCIES
|
||||
|
||||
Reference in New Issue
Block a user