vllm icon indicating copy to clipboard operation
vllm copied to clipboard

Published 20 hours ago • verified publisher icon vllm-project

[Feature]: Support HuggingFaceM4/idefics2-8b as vision model

Open pseudotensor opened this issue 1 year ago • 1 comments

🚀 The feature, motivation and pitch

I know even llava 1.6 isn't in, but benchmarks for this model are much better.

https://huggingface.co/HuggingFaceM4/idefics2-8b https://huggingface.co/blog/idefics2

image

TGI support coming according to model card discussion, but no PR in TGI yet for 2, but seems idefics is there.

Alternatives

llava1.6, but maybe obsolete by now.

Additional context

No response

pseudotensor avatar Apr 16 '24 22:04 pseudotensor

I will work on implementing this model.

jc9123 avatar Apr 18 '24 08:04 jc9123

I will work on implementing this model.

how is it going ,bro

xiechengmude avatar Jun 18 '24 13:06 xiechengmude

Any update here? Thanks.

dipta007 avatar Jul 12 '24 01:07 dipta007

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!

github-actions[bot] avatar Oct 29 '24 02:10 github-actions[bot]