tukaani-project/xz
XZ Utils
1,410
Generated Behavioral Tests
84.7%
Best Score
Claude Opus 4.7 (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 |  |
Claude Opus 4.7 (xhigh) | 84.7% | $27.85 | 344 | trace → |
| 2 |  |
GPT 5.5 (high) | 79.7% | $4.26 | 43 | trace → |
| 3 |  |
Gemini 3.5 Flash | 69.9% | $5.01 | 182 | trace → |
| 4 | |
GPT 5.5 (xhigh) | 44.7% | $12.08 | 82 | trace → |
| 5 | |
GPT 5.6 Sol (medium) | 44.7% | $2.66 | 23 | trace → |
| 6 | |
Claude Opus 4.8 (xhigh) | 41.4% | $17.38 | 108 | trace → |
| 7 | |
GPT 5.5 | 41.4% | $1.30 | 20 | trace → |
| 8 |  |
GLM-5.2 | 41.1% | $24.78 | 182 | trace → |
| 9 | |
Gemini 3.6 Flash | 39.4% | $2.55 | 142 | trace → |
| 10 | |
Claude Opus 4.7 | 36.0% | $3.27 | 94 | trace → |
| 11 | |
GPT 5.4 | 35.3% | $0.32 | 12 | trace → |
| 12 | |
Claude Haiku 4.5 | 30.1% | $0.45 | 119 | trace → |
| 13 | |
Gemini 3.1 Pro | 28.2% | $1.46 | 98 | trace → |
| 14 | |
GPT 5.4 mini | 26.5% | $0.02 | 8 | trace → |
| 15 | |
Gemini 3 Flash | 23.8% | $0.33 | 93 | trace → |
| 16 | |
GPT 5.6 Sol (xhigh) | 16.7% | $3.23 | 17 | trace → |
| 17 | |
Claude Sonnet 4.6 | 8.8% | $24.28 | 507 | trace → |
| 18 | |
GPT 5 mini | 5.6% | $0.06 | 18 | trace → |
| 19 | |
Claude Opus 4.6 | 0.0% | $1.36 | 189 | trace → |
Click a row to replay how that model rebuilt this program, or the model name to open its full run