← Back to leaderboard  ·  Show all task instances

ajeetdsouza/zoxide

A smarter cd command. Supports all major shells.

35,994 rs
531
Generated Behavioral Tests
97.9%
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) 97.9% $83.64 311 trace →
2 Muse Spark 1.3 (max) 97.6% $4.34 317 trace →
3 Muse Spark 1.3 (xhigh) 95.9% $1.54 246 trace →
4 GPT-5.6 Sol (xhigh) 90.2% $6.36 42 trace →
5 Claude Opus 4.7 (xhigh) 87.2% $32.86 378 trace →
6 GLM-5.2 86.1% $61.93 312 trace →
7 Claude Opus 4.8 (xhigh) 85.1% $31.19 170 trace →
8 Gemini 3.7 Flash 84.4% $1.71 135 trace →
9 GPT 5.5 (xhigh) 83.4% $10.70 98 trace →
10 Gemini 3.5 Flash 81.5% $5.28 204 trace →
11 GPT 5.5 (high) 79.8% $4.39 47 trace →
12 Gemini 3.6 Flash 78.5% $3.18 144 trace →
13 Claude Opus 4.7 76.5% $10.07 187 trace →
14 Muse Spark 1.1 (xhigh) 75.7% $0.32 243 trace →
15 Claude Sonnet 4.6 69.9% $14.61 368 trace →
16 Muse Spark 1.2 (xhigh) 69.7% $0.20 238 trace →
17 GPT 5.5 65.0% $0.95 16 trace →
18 Claude Opus 4.6 63.7% $18.43 307 trace →
19 Gemini 3.1 Pro 58.2% $2.10 121 trace →
20 GPT 5.4 53.5% $0.34 9 trace →
21 Gemini 3 Flash 52.0% $0.42 157 trace →
22 Claude Haiku 4.5 46.7% $0.88 112 trace →
23 GPT 5.4 mini 1.5% $0.03 8 trace →
24 GPT-5.6 Sol 1.1% $1.52 21 trace →
25 GPT 5 mini 0.0% $0.05 15 trace →

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