This is an adapter fine-tuned with QLoRA on top of Qwen2.5 72B Instruct quantized with bitsandbytes. It should only be used for testing QLoRA/LoRA. It has been fine-tuned only for 10 training steps.
- Downloads last month
- 5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support