jarun/nnn

n³ The unorthodox terminal file manager

21,506 c medium
477
Generated Behavioral Tests
98.5%
Best Score
Claude Opus 5 (xhigh)

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

21 runs
# Model Score Cost Calls
1 Claude Opus 5 (xhigh) 98.5% $78.92 381 trace →
2 GLM-5.2 98.3% $5.83 114 trace →
3 GPT 5.5 (high) 98.1% $3.95 36 trace →
4 Claude Opus 4.8 (xhigh) 98.1% $15.25 151 trace →
5 Claude Opus 4.6 98.1% $14.07 334 trace →
6 Claude Opus 4.7 (xhigh) 97.9% $1.88 70 trace →
7 Gemini 3.6 Flash 97.7% $2.43 89 trace →
8 GPT-5.6 Sol (xhigh) 96.9% $4.44 32 trace →
9 GPT-5.6 Sol 96.6% $0.63 11 trace →
10 Claude Opus 4.7 96.6% $1.32 56 trace →
11 Claude Sonnet 4.6 96.6% $10.47 347 trace →
12 GPT 5.5 96.4% $1.22 17 trace →
13 Gemini 3.7 Flash 91.4% $2.41 96 trace →
14 Claude Haiku 4.5 88.7% $0.52 123 trace →
15 Gemini 3.5 Flash 79.0% $3.53 99 trace →
16 GPT 5.4 76.7% $0.17 8 trace →
17 GPT 5.4 mini 69.8% $0.03 13 trace →
18 GPT 5.5 (xhigh) 64.8% $7.36 53 trace →
19 Gemini 3.1 Pro 59.7% $0.88 70 trace →
20 GPT 5 mini 51.4% $0.02 10 trace →
21 Gemini 3 Flash 13.8% $0.24 87 trace →

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