Is there any possibility to support `Phi-3-vision-128k-instruct`?

huggingface / text-generation-inference

Large Language Model Text Generation Inference

http://hf.co/docs/text-generation-inference

Apache License 2.0

8.73k stars 1.01k forks source link

Is there any possibility to support `Phi-3-vision-128k-instruct`? #2174

Closed farzanehnakhaee70 closed 3 weeks ago

farzanehnakhaee70 commented 2 months ago

Feature request

Microsoft published its light-weight version of multi-modal model which also supports images. Is there any possibility to support this model in TGI?

Motivation

Supporting multi-modal and scpecially models which will accept image as input in TGI.

Your contribution

No.

LysandreJik commented 1 month ago

Thanks for your request @farzanehnakhaee70! I'm making sure the team sees it.

For now the model is using custom code and can be used with transformers and trust_remote_code=True.

Usually what drives a model addition to our toolkit is the community excitement for that model to be added to our modeling libraries. Getting a lot of :+1: on your message will show us that it's important for a lot of people :)

github-actions[bot] commented 3 weeks ago

This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.