README.md

metadata

language:
  - en
license: apache-2.0
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - mistral
  - trl
  - sft
  - gguf
base_model: unsloth/mistral-7b-v0.2-bnb-4bit
datasets:
  - visheratin/realworldqa

Mistral-RealworldQA-7b SFT GGUF

Original fp16 weights can be found here.

An experiment with the goal of reducing halucinations in VQA. First in of a series of projects centering around fine-tuning for image captioning.

This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.