add flash attention support using triton #263
run_tests.yaml
on: push
tests on CPU with empty model
5m 50s
Artifacts
Produced during runtime
| Name | Size | Digest | |
|---|---|---|---|
|
htmlcov
Expired
|
2.09 MB |
sha256:f152448e1c39a4ace89ffcb0c61ba29e9190588ad742ca6f08cc627c71465b03
|
|
|
output
Expired
|
2.74 KB |
sha256:4de00d8a39a314a4772b27f84655bb2a32e012faba6753bffb70bd8ba295d31f
|
|