---
title: Medical Reasoning SFT 120B
emoji: 🏥
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.44.1
app_file: app.py
pinned: false
license: apache-2.0
---

# Medical Reasoning SFT 120B

Fine-tuned Llama 3.3 70B for medical reasoning using Adaption AutoScientist.

## Results

- Win Rate: 31 (base) -> 70 (adapted), +125.8%
- Science Win Rate: 22 (base) -> 78 (adapted), +254.5%
- Quality Score: 7.0 -> 7.6, +8.6%

## Links

- Dataset: https://huggingface.co/datasets/morningstarxcdcode/adaption-medical-reasoning-sft-120b
- Model: https://huggingface.co/morningstarxcdcode/adaption-medical-reasoning-sft-120b-model

Built for the AutoScientist Challenge by Adaption Labs.