CAIS Text Capabilities Index — leaderboard

Composite CAIS AI Dashboard text index averaging Humanity's Last Exam, ARC-AGI-2, TextQuests, and SWE-bench Pro for models with all component scores.

Metric: Text Capabilities Index (self-reported). Source: benchmarklist.com. Status: saturation imminent. 39 models tracked.

Top models

#ModelScore
1GPT-5.554.1
2Gemini 3.1 Pro (Preview)52.9
3GPT-5.449.3
4Gemini 3.5 Flash48.9
5Claude Opus 4.746.9
6Claude Opus 4.644
7Claude Opus 4.536.6
8Gemini 3 Flash (Preview)35.6
9GPT-5.233.8
10Claude Sonnet 4.632.6
11Grok 4.2032.5
12DeepSeek V4 Pro32.1
13GLM-5.129.8
14GPT-5.129
15Claude Sonnet 4.525.4

Interactive version: theaggregate.ai/benchmark?slug=cais-text-capabilities-index · How the rankings work · Data refreshed daily, snapshot 2026-07-22.