CoMMET (Text-Complete, With Images) - Belief and Desire: leaderboard
Metric: Story accuracy on the 21 second-order belief and desire StoryTurns among the 399 text-complete StoryTurns of CoMMET (all needed information is in the text), text with the illustrative images; story accuracy (%, a StoryTurn counts only if every question in the multi-turn conversation is answered correctly), open-ended answers judged against the ground truth by Gemini 3.1 Pro; higher is better. Source: arxiv.org. Saturation forecast: Around May 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 57.14 | #76 |
| 2 | Gemini 3.1 Flash Lite | 42.86 | #174 |
| 3 | Llama 4 Maverick | 38.1 | #451 |
| 4 | GPT-4o | 28.57 | #333 |
| 5 | Claude Opus 4.7 | 28.57 | #45 |
| 6 | Claude Haiku 4.5 | 14.29 | #271 |
| 7 | Llama 3.2 11B Instruct | 4.76 | #1112 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=commet-text-complete-with-images-belief-and-desire · How It Works · Data refreshed daily, snapshot 2026-10-11.