[Model Request] Mixtral-8x7B-v0.1

OpenGVLab / OmniQuant

[ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.

MIT License

626 stars 49 forks source link

[Model Request] Mixtral-8x7B-v0.1 #40

Closed joseph777111 closed 7 months ago

joseph777111 commented 7 months ago

The Mixtral-8x7B Large Language Model (LLM) is a pretrained generative Sparse Mixture of Experts. The Mistral-8x7B outperforms Llama 2 70B on most benchmarks we tested.

For full details of this model please read our release blog post.

https://huggingface.co/mistralai/Mixtral-8x7B-v0.1

ChenMnZ commented 7 months ago

Thanks for your proposal. I am working for this, please wait a moment.

ChenMnZ commented 7 months ago

We have supported Mixtral-8x7B-v0.1, please refer the latest code for more details. Some results based on lm-evaluation-harness is as follows:

joseph777111 commented 7 months ago

Thank you! You guys are freaking awesome!!!! 🤩