TurtleAI: leaderboard

Metric: Success rate (%) on all 823 TurtleAI-DS tasks (102 real, 102 hand-drawn, 619 synthetic), symbolic comparison: the predicted Python turtle program is run in a canonicalizing emulator and succeeds when its normalized drawing matches the reference pixel by pixel (0.95 or 0.92 overlap threshold), greedy decoding, one fixed prompt with the task image; higher is better. Source: arxiv.org. Saturation forecast: Around 2028. 18 models tracked.

Top models

#ModelScore
1O4 Mini (Medium)15.92
2O3 (Medium)13.73
3GPT-4o (2024-11-20)9.23
4GPT-5 (Medium)9.11
5Qwen 2.5 VL 72B Instruct6.8
6Qwen 3 VL 30B A3B Instruct6.2
7GPT-4 Turbo5.47
8InternVL3-78B4.98
9Pixtral-12B2.31
10Qwen 2 VL 7B Instruct0.12

Interactive version: theaggregate.ai/benchmark?slug=turtleai · How It Works · Data refreshed daily, snapshot 2026-09-29.