transformers
c24c79eb
- Optimize memory usage of mllama encoder (#34930)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 year ago
Optimize memory usage of mllama encoder (#34930) mllama encoder memory optimization
References
#34930 - Optimize memory usage of mllama encoder
#39821 - Support MetaCLIP 2
#59 - Fix attention mask handling in EoMT-DINOv3 converter
#62 - Add initial DEIMv2 model implementation
#65 - Fix RTDetrV2 sine position embedding ordering
Author
milesial
Parents
9ab8c5b5
Loading