From bd66b8529bb00821bf14565e8b7663cbdc286d91 Mon Sep 17 00:00:00 2001
From: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Date: Tue, 21 Oct 2025 15:23:56 +0100
Subject: [PATCH] [CI] Install pre-release version of `apache-tvm-ffi` for
 `flashinfer` (#27262)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
---
 docker/Dockerfile | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/docker/Dockerfile b/docker/Dockerfile
index 8f482b393c914..5d5b2a72b3dbb 100644
--- a/docker/Dockerfile
+++ b/docker/Dockerfile
@@ -132,7 +132,9 @@ WORKDIR /workspace
 COPY requirements/common.txt requirements/common.txt
 COPY requirements/cuda.txt requirements/cuda.txt
 RUN --mount=type=cache,target=/root/.cache/uv \
-    uv pip install --python /opt/venv/bin/python3 -r requirements/cuda.txt \
+    # TODO: remove apache-tvm-ffi once FlashInfer is fixed https://github.com/flashinfer-ai/flashinfer/issues/1962
+    uv pip install --python /opt/venv/bin/python3 --pre apache-tvm-ffi==0.1.0b15 \
+    && uv pip install --python /opt/venv/bin/python3 -r requirements/cuda.txt \
     --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')
 
 # cuda arch list used by torch
@@ -353,7 +355,9 @@ RUN --mount=type=cache,target=/root/.cache/uv \
 # Install vllm wheel first, so that torch etc will be installed.
 RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist \
     --mount=type=cache,target=/root/.cache/uv \
-    uv pip install --system dist/*.whl --verbose \
+    # TODO: remove apache-tvm-ffi once FlashInfer is fixed https://github.com/flashinfer-ai/flashinfer/issues/1962
+    uv pip install --system --pre apache-tvm-ffi==0.1.0b15 \
+    && uv pip install --system dist/*.whl --verbose \
         --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')
 
 # Install FlashInfer pre-compiled kernel cache and binaries