nachoparker/dutree
a tool to analyze file system usage written in Rust
641
Generated Behavioral Tests
97.5%
Best Score
GPT 5.6 Sol (xhigh)
Results by Model
Hover a point for details · The line marks the Pareto frontier (best score per cost) · Click a point to see model details
Per-Model Results
19 runs| # | Model | Score | Cost | Calls | ||
|---|---|---|---|---|---|---|
| 1 |  |
GPT 5.6 Sol (xhigh) | 97.5% | $4.69 | 32 | trace → |
| 2 | |
GPT 5.5 (xhigh) | 95.8% | $6.19 | 65 | trace → |
| 3 |  |
Gemini 3.6 Flash | 95.5% | $6.70 | 162 | trace → |
| 4 |  |
Claude Opus 4.8 (xhigh) | 94.5% | $19.22 | 112 | trace → |
| 5 | |
GPT 5.5 (high) | 94.1% | $3.96 | 36 | trace → |
| 6 | |
Claude Opus 4.6 | 89.2% | $13.09 | 243 | trace → |
| 7 |  |
GLM-5.2 | 88.1% | $10.42 | 145 | trace → |
| 8 | |
Claude Opus 4.7 (xhigh) | 86.7% | $19.52 | 133 | trace → |
| 9 | |
Gemini 3.5 Flash | 82.7% | $5.00 | 134 | trace → |
| 10 | |
GPT 5.5 | 78.8% | $1.14 | 17 | trace → |
| 11 | |
Claude Sonnet 4.6 | 78.0% | $17.62 | 335 | trace → |
| 12 | |
GPT 5.6 Sol (medium) | 67.2% | $0.96 | 16 | trace → |
| 13 | |
GPT 5.4 mini | 50.1% | $0.03 | 8 | trace → |
| 14 | |
GPT 5 mini | 44.5% | $0.01 | 8 | trace → |
| 15 | |
Claude Haiku 4.5 | 42.1% | $0.45 | 88 | trace → |
| 16 | |
Gemini 3 Flash | 38.5% | $0.33 | 81 | trace → |
| 17 | |
GPT 5.4 | 36.2% | $0.14 | 8 | trace → |
| 18 | |
Claude Opus 4.7 | 0.3% | $19.26 | 164 | trace → |
| 19 | |
Gemini 3.1 Pro | 0.0% | $0.69 | 54 | trace → |
Click a row to replay how that model rebuilt this program, or the model name to open its full run