← Back to leaderboard  ·  Show all task instances

mookid/diffr

Yet another diff highlighting tool

612 rs medium
606
Generated Behavioral Tests
93.1%
Best Score
Gemini 3.7 Flash

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 Gemini 3.7 Flash 93.1% $2.69 141 trace →
2 Claude Opus 5 (xhigh) 92.7% $80.33 261 trace →
3 Muse Spark 1.3 (xhigh) 92.1% $1.30 169 trace →
4 Claude Opus 4.8 (xhigh) 91.7% $26.68 134 trace →
5 GPT-5.6 Sol (xhigh) 91.4% $8.54 39 trace →
6 GPT 5.5 (xhigh) 90.6% $9.34 74 trace →
7 Muse Spark 1.2 (xhigh) 90.1% $1.43 133 trace →
8 Muse Spark 1.3 (max) 89.9% $9.46 380 trace →
9 Gemini 3.6 Flash 88.9% $15.26 182 trace →
10 Gemini 3.5 Flash 88.6% $11.93 193 trace →
11 GPT 5.5 (high) 88.4% $5.85 51 trace →
12 Claude Opus 4.7 (xhigh) 87.8% $27.18 193 trace →
13 GLM-5.2 87.5% $27.78 205 trace →
14 Claude Opus 4.7 84.7% $4.69 102 trace →
15 GPT-5.6 Sol 84.5% $1.05 15 trace →
16 Claude Sonnet 4.6 80.7% $27.33 384 trace →
17 Claude Opus 4.6 77.2% $13.03 208 trace →
18 GPT 5.5 74.8% $0.89 15 trace →
19 Gemini 3.1 Pro 66.2% $3.33 154 trace →
20 GPT 5.4 64.5% $0.19 8 trace →
21 GPT 5 mini 55.1% $0.03 15 trace →
22 Claude Haiku 4.5 54.6% $1.11 143 trace →
23 Gemini 3 Flash 52.3% $0.25 76 trace →
24 Muse Spark 1.1 (xhigh) 1.2% $1.19 270 trace →
25 GPT 5.4 mini 0.0% $0.02 7 trace →

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