Update Flashinfer to 0.6.6 (#36768)

Signed-off-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>
Co-authored-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>

docker/Dockerfile

@@ -586,7 +586,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
 # This is ~1.1GB and only changes when FlashInfer version bumps
 # https://docs.flashinfer.ai/installation.html
 # From versions.json: .flashinfer.version
-ARG FLASHINFER_VERSION=0.6.4
+ARG FLASHINFER_VERSION=0.6.6
 RUN --mount=type=cache,target=/root/.cache/uv \
     uv pip install --system flashinfer-cubin==${FLASHINFER_VERSION} \
     && uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} \

docker/Dockerfile.nightly_torch

@@ -217,13 +217,13 @@ RUN pip install setuptools==75.6.0 packaging==23.2 ninja==1.11.1.3 build==1.2.2.
 
 
 # build flashinfer for torch nightly from source around 10 mins
-# release version: v0.6.4
+# release version: v0.6.6
 # todo(elainewy): cache flashinfer build result for faster build
 ENV CCACHE_DIR=/root/.cache/ccache
 RUN --mount=type=cache,target=/root/.cache/ccache \
     --mount=type=cache,target=/root/.cache/uv \
     echo "git clone flashinfer..." \
-    && git clone --depth 1 --branch v0.6.4 --recursive https://github.com/flashinfer-ai/flashinfer.git \
+    && git clone --depth 1 --branch v0.6.6 --recursive https://github.com/flashinfer-ai/flashinfer.git \
     && cd flashinfer \
     && git submodule update --init --recursive \
     && echo "finish git clone flashinfer..." \

docker/versions.json

@@ -65,7 +65,7 @@
       "default": "true"
     },
     "FLASHINFER_VERSION": {
-      "default": "0.6.4"
+      "default": "0.6.6"
     },
     "GDRCOPY_CUDA_VERSION": {
       "default": "12.8"

requirements/cuda.txt

@@ -9,7 +9,7 @@ torchaudio==2.10.0
 # These must be updated alongside torch
 torchvision==0.25.0 # Required for phi3v processor. See https://github.com/pytorch/vision?tab=readme-ov-file#installation for corresponding version
 # FlashInfer should be updated together with the Dockerfile
-flashinfer-python==0.6.4
+flashinfer-python==0.6.6
 # Cap nvidia-cudnn-frontend (transitive dep of flashinfer) due to
 # breaking changes in 1.19.0
 nvidia-cudnn-frontend>=1.13.0,<1.19.0

tools/pre_commit/update-dockerfile-graph.sh

@@ -41,7 +41,7 @@ if printf '%s\n' "${FILES[@]}" | grep -q "^docker/Dockerfile$"; then
     --rm \
     --user "$(id -u):$(id -g)" \
     --workdir /workspace \
-    --volume "$(pwd)":/workspace \
+    --volume "$(pwd -P)":/workspace \
     ghcr.io/patrickhoefler/dockerfilegraph:alpine \
     --output png \
     --dpi 200 \