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
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 54.1 |
| 2 | Gemini 3.1 Pro (Preview) | 52.9 |
| 3 | GPT-5.4 | 49.3 |
| 4 | Gemini 3.5 Flash | 48.9 |
| 5 | Claude Opus 4.7 | 46.9 |
| 6 | Claude Opus 4.6 | 44 |
| 7 | Claude Opus 4.5 | 36.6 |
| 8 | Gemini 3 Flash (Preview) | 35.6 |
| 9 | GPT-5.2 | 33.8 |
| 10 | Claude Sonnet 4.6 | 32.6 |
| 11 | Grok 4.20 | 32.5 |
| 12 | DeepSeek V4 Pro | 32.1 |
| 13 | GLM-5.1 | 29.8 |
| 14 | GPT-5.1 | 29 |
| 15 | Claude Sonnet 4.5 | 25.4 |
Interactive version: theaggregate.ai/benchmark?slug=cais-text-capabilities-index · How the rankings work · Data refreshed daily, snapshot 2026-07-22.