---
base_model: mistralai/Mistral-7B-Instruct-v0.2
---

Convert from mistralai/Mistral-7B-Instruct-v0.2, and 4 bits quantized.