← Back to leaderboard  ·  Show all task instances

noborus/ov

🎑Feature-rich terminal-based text viewer. It is a so-called terminal pager.

1,935 go medium
1,802
Generated Behavioral Tests
94.7%
Best Score
Claude Opus 5 (xhigh)

Results by Model

Hover for details · Click a model to open it · Drag to zoom the x-axis · Shift-drag to pan · The line marks the Pareto frontier (best score per cost)

Per-Model Results

25 runs
# Model Score Cost Calls
1 Claude Opus 5 (xhigh) 94.7% $21.33 174 trace →
2 Muse Spark 1.3 (max) 91.5% $5.19 303 trace →
3 Claude Opus 4.6 89.4% $12.67 313 trace →
4 Claude Opus 4.8 (xhigh) 88.8% $9.40 102 trace →
5 Claude Sonnet 4.6 88.8% $13.13 389 trace →
6 GLM-5.2 88.6% $15.12 214 trace →
7 Gemini 3.7 Flash 88.2% $3.05 159 trace →
8 GPT 5.5 (high) 86.9% $3.23 41 trace →
9 Muse Spark 1.2 (xhigh) 86.2% $0.43 111 trace →
10 GPT-5.6 Sol (xhigh) 85.0% $5.57 43 trace →
11 Gemini 3.6 Flash 84.0% $4.34 151 trace →
12 Gemini 3.5 Flash 83.9% $4.04 133 trace →
13 GPT 5.5 (xhigh) 81.6% $10.99 88 trace →
14 GPT-5.6 Sol 80.1% $1.11 14 trace →
15 Muse Spark 1.3 (xhigh) 74.5% $0.91 190 trace →
16 GPT 5.5 74.5% $1.55 22 trace →
17 Muse Spark 1.1 (xhigh) 74.2% $1.04 207 trace →
18 Claude Opus 4.7 73.6% $2.35 99 trace →
19 Claude Opus 4.7 (xhigh) 71.5% $4.49 136 trace →
20 Claude Haiku 4.5 58.9% $0.48 97 trace →
21 GPT 5.4 38.6% $0.21 11 trace →
22 GPT 5 mini 37.3% $0.01 10 trace →
23 GPT 5.4 mini 11.5% $0.03 28 trace →
24 Gemini 3 Flash 4.2% $0.19 62 trace →
25 Gemini 3.1 Pro 2.1% $2.37 111 trace →

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