transformers
913330ca
- VLMs: fix number of image tokens (#34332)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 year ago
VLMs: fix number of image tokens (#34332) * fix * fix tests * add tests * style * style * fix qwen after rebase * fix video llava
References
#62 - Add initial DEIMv2 model implementation
#58 - Add EoMT DINOv3 model
#59 - Fix attention mask handling in EoMT-DINOv3 converter
#41212 - Add EoMT with DINOv3 backbone
#39821 - Support MetaCLIP 2
#34332 - VLMs: fix number of image tokens
Author
zucchini-nlp
Parents
0f764a5a
Loading