data/eval_report_3b_v4.md

# CodeWraith Model Evaluation Report

## Summary

| Metric | CodeWraith-3b (Llama-3.2-3B-Instruct) |
|--------|-----|
| Avg Structural Score | 0.91 |
| Function Coverage | 0.86 |
| Class Coverage | 0.96 |
| Argument Coverage | 0.93 |
| Return Type Coverage | 0.67 |
| Good Scores (>=80%) | 24 |
| Avg Inference Time (s) | 25.57 |

## CodeWraith-3b (Llama-3.2-3B-Instruct)

- Examples evaluated: 28
- Valid (parseable): 27
- Perfect scores: 19
- Total inference time: 715.9s