OmniaBench - Program: leaderboard

Metric: Pass@1 (%; 30 procedural tasks with branching and iteration verified by executable VerifyCode checks; a task counts as solved only when every rubric item is satisfied (Program tasks by a VerifyCode check); up to 200 tool-call steps; challenging set of 644 tasks evaluated June 29 to July 15, 2026). Source: arxiv.org. Saturation forecast: Around December 2026. 22 models tracked.

Top models

#ModelScore
1Qwen 3.7 Max (Thinking)66.67
2Kimi K2.6 (Thinking)63.33
3Claude Opus 4.7 (Max)63.33
4Claude Sonnet 5 (Max)63.33
5GLM-5.2 (Max)60
6GPT-5.5 (High)60
7DeepSeek V4 Pro (Max)53.33
8GLM-5 (Thinking)53.33
9Qwen 3.5 397B A17B50
10GPT-5.6 Sol (High)50
11Qwen 3.6 27B46.67
12DeepSeek V4 Flash (Max)43.33
13Qwen 3.6 35B A3B36.67
14Gemini 3.1 Pro (Preview)30
15Qwen 3.5 35B A3B30

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