Commit History

Autor SHA1 Mensaxe Data
  Matthias Reso e2f77dbc21 fix quant config hai 9 meses
  Matthias Reso 6ef9a78458 Fix issues with quantization_config == None hai 9 meses
  Matthias Reso 0920b1a415 Fix quantization for inference hai 9 meses
  Hamid Shojanazeri d51d2cce9c adding sdpa for flash attn hai 1 ano
  Hamid Shojanazeri db8af96ff0 update the model load with native flash attn hai 1 ano
  Matthias Reso 4c9cc7d223 Move modules into separate src folder hai 1 ano