Skip to content

fix(qwen-tts): install flash-attn on CUDA 13 #11251

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
localai-org-maint-bot wants to merge 4 commits into
base: master
Choose a base branch
from
+17 −0
Open

fix(qwen-tts): install flash-attn on CUDA 13 #11251

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
41d4251
fix(qwen-tts): install flash-attn on CUDA 13
localai-org-maint-bot Jul 31, 2026
abb6e2b
fix(qwen-tts): serialize CUDA 13 flash-attn build
localai-org-maint-bot Aug 1, 2026
7916c48
test(qwen-tts): guard CUDA 13 build concurrency
localai-org-maint-bot Aug 1, 2026
127a578
fix(qwen-tts): limit CUDA compiler threads
localai-org-maint-bot Aug 1, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions backend/python/qwen-tts/install.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,4 +10,11 @@ else
source $backend_dir/../common/libbackend.sh
fi

# CUDA 13 has no prebuilt FlashAttention wheel, so the fallback source build
# exceeds the CI runner's memory when ninja compiles multiple units at once.
if [ "x${BUILD_PROFILE}" = "xcublas13" ]; then
export MAX_JOBS="${MAX_JOBS:-1}"
export NVCC_THREADS="${NVCC_THREADS:-1}"
fi

installRequirements
1 change: 1 addition & 0 deletions backend/python/qwen-tts/requirements-cublas13-after.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
flash-attn
9 changes: 9 additions & 0 deletions backend/python/qwen-tts/test.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,15 @@
set -e

backend_dir=$(dirname $0)

for cuda_version in 12 13; do
grep -qx "flash-attn" "$backend_dir/requirements-cublas${cuda_version}-after.txt"
done

grep -q 'BUILD_PROFILE.*cublas13' "$backend_dir/install.sh"
grep -q 'MAX_JOBS.*1' "$backend_dir/install.sh"
grep -q 'NVCC_THREADS.*1' "$backend_dir/install.sh"

if [ -d $backend_dir/common ]; then
source $backend_dir/common/libbackend.sh
else
Expand Down
Loading