flashinfer-ai / flashinfer

FlashInfer: Kernel Library for LLM Serving
https://flashinfer.ai
Apache License 2.0
1.46k stars 143 forks source link

bugfix: fix prefill kernel uris for aot compilation #624

Closed yzh119 closed 3 days ago

yzh119 commented 3 days ago

mask is no longer part of uris, this PR fixes the issue, otherwise our aot wheels will still trigger JIT compilation for prefill kernels.