---
license: apache-2.0
---

EXL2 quants of [Mistral-7B-instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3/edit/main/README.md)

v0.3's vocabulary is compatible with Mistral-Large-123B, so this works as a draft model for Mistral-Large.

[2.80 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/2.8bpw)    
[3.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/3.0bpw)    
[4.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/4.0bpw)    
[4.50 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/4.5bpw)    
[5.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/5.0bpw)    
[6.00 bits per weight](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/tree/6.0bpw)    

[measurement.json](https://huggingface.co/turboderp/Mistral-7B-instruct-v0.3-exl2/blob/main/measurement.json)