Update makefile

Signed-off-by: SimJeg <sjegou@nvidia.com>

Author: SimJeg

.github/workflows/test.yml

@@ -29,6 +29,12 @@ jobs:
         with:
           enable-cache: true
 
+      - name: Cache flash-attn build
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pip
+          key: flash-attn-${{ runner.os }}-cuda12.5-py3.10-${{ hashFiles('uv.lock') }}
+
       - name: Install dependencies
         run: uv sync --all-groups
 

Makefile

@@ -42,7 +42,7 @@ reports:
 .PHONY: test
 test: reports
 	$(UV) pip install optimum-quanto
-	$(UV) pip install flash-attn --find-links https://github.com/Dao-AILab/flash-attention/releases/expanded_assets/v2.8.3
+	$(UV) pip install flash-attn --no-build-isolation
 	PYTHONPATH=. \
 	$(UV) run pytest \
 		--cov-report xml:reports/coverage.xml \