Hugging Face's logo

google
/
translategemma-27b-it

Image-Text-to-Text
Transformers
Safetensors
gemma3
conversational
text-generation-inference
Model card Files
xet
Community
5

MoE or E3B | E4B

#4
by E7Reine - opened
Discussion
E7Reine

Are you planning to make MoE or E3B | E4B models, for example: 24 or "27B E3B" ?
they would be faster than dense models
sonali-kumari11
 Google org

Hi @E7Reine ,

Thank you for reaching out to us.

We appreciate your interest in MoE and E3B/E4B architectures. Your request for variants like E3B and MoE has been shared with the development team. Thank you for the feedback.
E7Reine

Hi @E7Reine ,

Thank you for reaching out to us.

We appreciate your interest in MoE and E3B/E4B architectures. Your request for variants like E3B and MoE has been shared with the development team. Thank you for the feedback.

Thanks! I'm looking forward to the release of these models.

Sign up or log in to comment