Conceptual Reasoning Index - Decision Theory (DTBench): leaderboard
Metric: Chance-Corrected Score (0-100). Source: conceptualreasoning.ai. 192 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 97.33 |
| 2 | Claude Opus 5 | 96 |
| 3 | Grok 4.6 | 95.55 |
| 4 | Gemini 3.7 Flash | 94.67 |
| 5 | Grok 4.5 | 94.22 |
| 6 | GPT-5.5 | 93.33 |
| 7 | GPT-5.5 Pro | 93.33 |
| 8 | GPT-5.6 Pro Sol | 93.33 |
| 9 | Gemini 3.1 Pro (Preview) | 92.45 |
| 10 | GPT-5.6 Sol | 92.45 |
| 11 | Gemini 3.6 Flash | 92.45 |
| 12 | Claude Opus 4.8 | 91.55 |
| 13 | Gemini 3.5 Flash | 91.12 |
| 14 | Muse Spark 1.2 | 91.12 |
| 15 | Claude Opus 4.7 | 91.08 |
Interactive version: theaggregate.ai/benchmark?slug=conceptual-reasoning-index-decision-theory-dtbench · How It Works · Data refreshed daily, snapshot 2026-09-05.