Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BOUNTY - $100] Add support for LLaVA #3

Closed
AlexCheema opened this issue Jul 15, 2024 · 17 comments
Closed

[BOUNTY - $100] Add support for LLaVA #3

AlexCheema opened this issue Jul 15, 2024 · 17 comments
Labels
good first issue Good for newcomers

Comments

@AlexCheema
Copy link
Contributor

Motivation: Vision models are useful.

What: exo supports different inference engines. Choose an inference engine (currently only MLX and tinygrad) for your implementation. You can probably find an existing LLaVA implementation somewhere (perhaps in https://github.com/ml-explore/mlx-examples for MLX). Modify that implementation to support sharded inference.

Reward: $100 Bounty paid out with USDC on Ethereum, email [email protected].

@AlexCheema AlexCheema changed the title [BOUNTY] Add support for LLaVA [BOUNTY - $100] Add support for LLaVA Jul 15, 2024
@anirishrnbsinger
Copy link

taking a crack at this

@maciejjankowski
Copy link

maciejjankowski commented Jul 16, 2024

taking a crack at this

Pair programming? (Keep the bounty)

@anirishrnbsinger
Copy link

anirishrnbsinger commented Jul 17, 2024 via email

@AlexCheema AlexCheema added the good first issue Good for newcomers label Jul 18, 2024
@AlexCheema
Copy link
Contributor Author

go for it guys. just leave a comment here if you need any help or run into issues :) @maciejjankowski @anirishrnbsinger

@anirishrnbsinger
Copy link

Didn't make as much progress as would've hoped. Will be backed up with work for a couple days. Anyone else welcome to take a crack before then

@INF800
Copy link

INF800 commented Jul 21, 2024

Interesting stuff. Wanted to read more about this but docs seem to be WIP.

https://github.com/exo-explore/exo?tab=readme-ov-file#documentation

@INF800
Copy link

INF800 commented Jul 21, 2024

Ain't this as straightforward as using the code from exo llama example and using it with mlx llava example?

@INF800
Copy link

INF800 commented Jul 21, 2024

I can do pair programming, planning to do it on weekend 28th :)

cc: @maciejjankowski

@INF800
Copy link

INF800 commented Jul 21, 2024

Ain't this as straightforward as using the code from exo llama example and using it with mlx llava example?

cc: @AlexCheema

And when can we expect the docs

@abhishek818
Copy link

attempting this..

@abhishek818
Copy link

@AlexCheema hey, seems exo is not yet supported for windows platform? Is there any plan to support it in future.

@khushgx
Copy link

khushgx commented Jul 26, 2024

taking a crack at this

if yall still need help, would love to get involved. (Also do not care abt bounty and u can keep it)

@AlexCheema
Copy link
Contributor Author

AlexCheema commented Jul 27, 2024

taking a crack at this

if yall still need help, would love to get involved. (Also do not care abt bounty and u can keep it)

@khushgx looks like @varshith15 has started on it here: #88
You could jump in there to support!

@varshith15
Copy link
Contributor

varshith15 commented Jul 27, 2024

im almost done sharding the model and removing duplicate code
if anyone wants to work on it, please pick up the client part of the code (UI changes) and the code connecting it to the model @khushgx

@AshirMehmood
Copy link

AshirMehmood commented Aug 1, 2024

is there any thing I can contribute in ? @varshith15

@AlexCheema
Copy link
Contributor Author

This is completed. Thanks @varshith15 don’t forget to shoot me an email [email protected] to claim your bounty payout.

@varshith15
Copy link
Contributor

@AlexCheema emailed you a couple of days ago, [email protected]
Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
good first issue Good for newcomers
Projects
None yet
Development

No branches or pull requests

8 participants