Hover a point for details · The line marks the Pareto frontier (best score per cost) · Click a point to see model details

19 runs
# Model Score Cost Calls
1 Claude Opus 4.8 (xhigh) 85.4% $34.95 188 trace →
2 GPT 5.5 (xhigh) 80.9% $10.29 95 trace →
3 GPT 5.6 Sol (xhigh) 80.4% $5.90 36 trace →
4 GPT 5.5 (high) 77.2% $4.32 48 trace →
5 Claude Opus 4.7 (xhigh) 76.6% $27.07 238 trace →
6 Claude Opus 4.6 64.1% $15.75 278 trace →
7 Gemini 3.5 Flash 62.8% $5.90 190 trace →
8 Claude Opus 4.7 58.9% $7.97 178 trace →
9 GPT 5.5 56.3% $1.11 16 trace →
10 Claude Sonnet 4.6 53.8% $36.37 584 trace →
11 GPT 5.6 Sol (medium) 53.1% $0.90 16 trace →
12 Gemini 3 Flash 38.7% $0.40 92 trace →
13 GPT 5.4 27.7% $0.28 7 trace →
14 GLM-5.2 27.5% $11.84 146 trace →
15 Claude Haiku 4.5 16.6% $0.81 135 trace →
16 GPT 5 mini 12.5% $0.03 16 trace →
17 Gemini 3.6 Flash 0.0% $12.22 263 trace →
18 Gemini 3.1 Pro 0.0% $1.30 76 trace →
19 GPT 5.4 mini 0.0% $0.02 8 trace →

Click a row to replay how that model rebuilt this program, or the model name to open its full run