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