# Phonton Plan Benchmark

- Generated: 2026-05-03T19:28:39.3846267+07:00
- Commit: 5f3c635
- Version: phonton 0.3.0
- Benchmark type: plan-preview-token-estimate

> These are planner estimates, not provider billing records. Use them as release evidence, not as public proof of end-to-end token savings.

| Goal | Status | Subtasks | Phonton estimate | Naive baseline | Estimated reduction | Time |
|---|---:|---:|---:|---:|---:|---:|
| add input validation to config loading | pass | 1 | 3500 | 35000 | 90% | 783ms |
| improve provider auth error messages | pass | 1 | 3500 | 35000 | 90% | 475ms |
| write tests for rollback failure handling | pass | 1 | 3500 | 35000 | 90% | 474ms |
| add a review summary for verified diffs | pass | 1 | 3500 | 35000 | 90% | 480ms |

Raw JSON: `benchmarks\results\plan-benchmark-20260503-192834.json`