Nimbus Coffee Assistant (0.5B)
A fine-tuned Qwen 2.5 0.5B model trained to answer customer support questions for Nimbus Coffee Roasters.
Model Details
- Base Model: Qwen/Qwen2.5-0.5B
- Fine-tuning Method: LoRA
- Dataset: 80 Q&A pairs covering products, pricing, shipping, and policies
- Epochs: 25
- Final Loss: 0.10
Training Progression
| Model | Dataset | Epochs | Final Loss |
|---|
| 0.5B | 20 examples | 5 | 1.32 |
| 1.5B | 50 examples | 15 | 0.24 |
| 3B | 80 examples | 25 | 0.04 |
| 0.5B (this model) | 80 examples | 25 | 0.10 |
Intended Use
This model is designed for live deployment on resource-constrained platforms. It demonstrates that a small fine-tuned model can deliver accurate domain-specific responses.
Limitations
- Trained on a small dataset (80 examples)
- May not handle out-of-scope questions well
- Designed for demonstration, not production use
Deployment
Deployed as a Streamlit app using this model for low-latency inference.
