OmniaBench — leaderboard

Metric: Overall Pass@1 (%). Source: scuuy.github.io. 22 models tracked.

Top models

#ModelScore
1Claude Sonnet 558.54
2GPT-5.6 Sol57.14
3GLM-5.256.83
4GPT-5.556.52
5DeepSeek V4 Pro54.5
6Claude Opus 4.754.19
7Kimi K2.6 (Thinking)52.33
8Qwen 3.7 Max49.69
9DeepSeek V4 Flash48.6
10Gemini 3.5 Flash45.65
11GLM-5 (Thinking)43.32
12Seed 2.1 Pro42.7
13Qwen 3.6 27B42.08
14Gemini 3.1 Pro (Preview)41.3
15Qwen 3.5 397B A17B41.15

Interactive version: aibenchmarks.dev/benchmark?slug=omniabench · How the rankings work · Data refreshed daily, snapshot 2026-07-19.