You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It'd be nice for ctranslate2 to have compatibility with at least one model that has vision model capabilities. since ct2 already supports gemma, it'd be a good opportunity to add one since gemma3 has a vision component:
see e.g.
google/gemma-3-12b-it
The text was updated successfully, but these errors were encountered:
Yes, it will be nice if you add support of Gemma3 conversion. It has completely different architecture (due to multimodal), but the text part of the model is pretty similar to Gemma2.
Also Qwen3 models also appeared and they are waiting for CTranslate2 :)
It'd be nice for ctranslate2 to have compatibility with at least one model that has vision model capabilities. since ct2 already supports gemma, it'd be a good opportunity to add one since gemma3 has a vision component:
see e.g.
google/gemma-3-12b-it
The text was updated successfully, but these errors were encountered: