mlc-ai / web-llm

High-performance In-browser LLM Inference Engine

Home Page:https://webllm.mlc.ai

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Different instruct based models

nico-martin opened this issue · comments

Hi there,
Did anyone already use other instruction-tuned models than the Mistral 7b?
Maybe like the RedPajama 3b or 7b instruct?
https://huggingface.co/togethercomputer/RedPajama-INCITE-Instruct-3B-v1
https://huggingface.co/togethercomputer/RedPajama-INCITE-7B-Instruct

I think Gemma 2b and Gemma 7b would be super interresting as well.

Just added Gemma 2b to https://webllm.mlc.ai/, 7b should be supported soon!

For anyone wondering, this tutorial may be relevant for adding model variants (i.e. same model library, just different weights): https://llm.mlc.ai/docs/deploy/javascript.html#bring-your-own-model-variant