diff --git a/docker/Dockerfile b/docker/Dockerfile index b2c949d2d1..550cc6f4dc 100644 --- a/docker/Dockerfile +++ b/docker/Dockerfile @@ -758,7 +758,7 @@ RUN --mount=type=cache,target=/opt/uv/cache \ # Install FlashInfer JIT cache (requires CUDA-version-specific index URL) # https://docs.flashinfer.ai/installation.html # From versions.json: .flashinfer.version -ARG FLASHINFER_VERSION=0.6.16.post3 +ARG FLASHINFER_VERSION=0.6.17 RUN --mount=type=cache,target=/opt/uv/cache \ uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} \ --index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') diff --git a/docker/versions.json b/docker/versions.json index ea51010c90..7ac2bada5e 100644 --- a/docker/versions.json +++ b/docker/versions.json @@ -68,7 +68,7 @@ "default": "true" }, "FLASHINFER_VERSION": { - "default": "0.6.16.post3" + "default": "0.6.17" }, "GDRCOPY_CUDA_VERSION": { "default": "12.8" diff --git a/requirements/cuda.txt b/requirements/cuda.txt index 3aa9016fd8..530c223bea 100644 --- a/requirements/cuda.txt +++ b/requirements/cuda.txt @@ -14,8 +14,8 @@ PyNvVideoCodec==2.0.4 # flashinfer-cubin is not on PyPI since 0.6.14; setup.py excludes it from # install_requires so the published wheel does not carry an unresolvable pin --extra-index-url https://flashinfer.ai/whl/ -flashinfer-python==0.6.16.post3 -flashinfer-cubin==0.6.16.post3 +flashinfer-python==0.6.17 +flashinfer-cubin==0.6.17 apache-tvm-ffi==0.1.11 tilelang==0.1.12 nvidia-cudnn-frontend>=1.19.1