onnxruntime
2a911505
- Optimize fp16 attention bias application
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
192 days ago
Optimize fp16 attention bias application
References
derdeljan/optimize_16bit_gqa
Author
derdeljan-msft
Parents
27f4dcd4
Loading