[GH-ISSUE #2970] feat: browser wasm llm #13087

Closed
opened 2026-04-19 19:54:40 -05:00 by GiteaMirror · 6 comments
Owner

Originally created by @tjbck on GitHub (Jun 10, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/2970

Originally created by @tjbck on GitHub (Jun 10, 2024). Original GitHub issue: https://github.com/open-webui/open-webui/issues/2970
Author
Owner

@justinh-rahb commented on GitHub (Jun 10, 2024):

https://github.com/mlc-ai/web-llm or something like it?

<!-- gh-comment-id:2157174760 --> @justinh-rahb commented on GitHub (Jun 10, 2024): https://github.com/mlc-ai/web-llm or something like it?
Author
Owner

@lukestanley commented on GitHub (Jun 12, 2024):

This is useful if you can only do a limited amount of processing on the server, and where the client has more processing power. WASM approaches to running Llama.cpp exist, as well as web GPU approaches (that only work on certain systems and and certain browser versions). There may be a WASM memory size issue with normal LLM sizes though. Working around it may well be possible though.

<!-- gh-comment-id:2162603590 --> @lukestanley commented on GitHub (Jun 12, 2024): This is useful if you can only do a limited amount of processing on the server, and where the client has more processing power. WASM approaches to running Llama.cpp exist, as well as web GPU approaches (that only work on certain systems and and certain browser versions). There may be a WASM memory size issue with normal LLM sizes though. Working around it may well be possible though.
Author
Owner

@tjbck commented on GitHub (Jun 13, 2024):

@justinh-rahb yep, seems promising!

<!-- gh-comment-id:2166898092 --> @tjbck commented on GitHub (Jun 13, 2024): @justinh-rahb yep, seems promising!
Author
Owner

@BuildBackBuehler commented on GitHub (Jun 21, 2024):

@justinh-rahb yep, seems promising!

Yeah MLC-LLM is the ship!
https://github.com/open-webui/open-webui/issues/1270

Would love to plug-n-play with it

<!-- gh-comment-id:2183558809 --> @BuildBackBuehler commented on GitHub (Jun 21, 2024): > @justinh-rahb yep, seems promising! Yeah MLC-LLM is the ship! https://github.com/open-webui/open-webui/issues/1270 Would love to plug-n-play with it
Author
Owner

@sala91 commented on GitHub (Jul 28, 2024):

WASM multi-threading is not good to this day, having been developing on top of it for years. This could work for smaller models tho, e.g. running a tiny model to translate between languages on the fly for better model output.

<!-- gh-comment-id:2254462558 --> @sala91 commented on GitHub (Jul 28, 2024): WASM multi-threading is not good to this day, having been developing on top of it for years. This could work for smaller models tho, e.g. running a tiny model to translate between languages on the fly for better model output.
Author
Owner

@justinh-rahb commented on GitHub (Jul 28, 2024):

This is new and potentially useful as well:

https://developer.chrome.com/docs/ai/built-in

<!-- gh-comment-id:2254692560 --> @justinh-rahb commented on GitHub (Jul 28, 2024): This is new and potentially useful as well: https://developer.chrome.com/docs/ai/built-in
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#13087