← Back to leaderboard  ·  Show all task instances

ast-grep/ast-grep

⚡A CLI tool for code structural search, lint and rewriting. Written in Rust

13,541 rs
877
Generated Behavioral Tests
57.2%
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) 57.2% $42.34 228 trace →
2 Muse Spark 1.3 (max) 45.2% $3.96 358 trace →
3 GPT-5.6 Sol (xhigh) 32.7% $5.19 39 trace →
4 Muse Spark 1.3 (xhigh) 28.2% $1.67 293 trace →
5 Gemini 3.7 Flash 27.6% $1.38 141 trace →
6 Gemini 3.6 Flash 24.5% $3.30 143 trace →
7 GLM-5.2 23.9% $24.06 210 trace →
8 Muse Spark 1.1 (xhigh) 23.8% $0.14 157 trace →
9 GPT 5.5 (xhigh) 23.7% $8.63 67 trace →
10 GPT 5.5 (high) 20.3% $3.93 35 trace →
11 Muse Spark 1.2 (xhigh) 18.7% $0.12 143 trace →
12 Claude Opus 4.8 (xhigh) 17.3% $13.83 116 trace →
13 GPT 5.5 12.3% $1.48 21 trace →
14 Claude Opus 4.6 11.7% $11.94 322 trace →
15 GPT-5.6 Sol 10.6% $0.73 11 trace →
16 Claude Opus 4.7 (xhigh) 9.6% $5.83 123 trace →
17 GPT 5.4 9.2% $0.41 11 trace →
18 Claude Sonnet 4.6 8.1% $28.64 621 trace →
19 Gemini 3.5 Flash 7.8% $6.16 138 trace →
20 Gemini 3.1 Pro 5.8% $0.79 66 trace →
21 Claude Opus 4.7 3.9% $1.45 46 trace →
22 Gemini 3 Flash 2.6% $0.27 70 trace →
23 Claude Haiku 4.5 2.6% $0.49 92 trace →
24 GPT 5 mini 1.4% $0.04 16 trace →
25 GPT 5.4 mini 0.0% $0.05 38 trace →

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