This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B 1 with Unsloth.

It was trained on 100k samples from the lavita/ChatDoctor-HealthCareMagic-100k dataset.

This model was made only for demonstration purposes.

This model is not suitable for real-world medical advice or diagnosis. Users should not rely on it for clinical decision-making or patient care.

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