← Back to leaderboard  ·  Show all task instances

rust-embedded/svd2rust

Generate Rust register maps (`struct`s) from SVD files

835 rs medium
918
Generated Behavioral Tests
93.2%
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) 93.2% $93.27 429 trace →
2 Muse Spark 1.3 (xhigh) 83.6% $2.84 308 trace →
3 GPT 5.5 (xhigh) 81.6% $21.57 127 trace →
4 Muse Spark 1.3 (max) 77.1% $14.84 471 trace →
5 GLM-5.2 75.3% $53.01 291 trace →
6 Claude Sonnet 4.6 72.7% $25.43 389 trace →
7 Muse Spark 1.2 (xhigh) 72.4% $2.17 176 trace →
8 Claude Opus 4.8 (xhigh) 72.3% $17.07 137 trace →
9 Gemini 3.6 Flash 69.4% $3.84 121 trace →
10 GPT 5.5 (high) 68.6% $4.94 39 trace →
11 Claude Opus 4.6 65.6% $12.91 218 trace →
12 GPT 5.5 61.4% $1.83 28 trace →
13 Claude Opus 4.7 (xhigh) 57.7% $1.81 50 trace →
14 GPT-5.6 Sol 55.9% $0.61 11 trace →
15 Gemini 3.1 Pro 45.3% $1.86 117 trace →
16 Claude Haiku 4.5 30.5% $0.66 92 trace →
17 GPT-5.6 Sol (xhigh) 15.9% $13.54 75 trace →
18 Claude Opus 4.7 14.5% $0.61 27 trace →
19 GPT 5.4 9.4% $0.09 6 trace →
20 GPT 5.4 mini 8.0% $0.03 9 trace →
21 GPT 5 mini 7.2% $0.04 19 trace →
22 Gemini 3.7 Flash 4.4% $1.12 94 trace →
23 Muse Spark 1.1 (xhigh) 0.8% $0.41 141 trace →
24 Gemini 3 Flash 0.8% $0.39 82 trace →
25 Gemini 3.5 Flash 0.0% $11.97 149 trace →

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