MixRea - Implicit Information Accuracy: leaderboard
Metric: Implicit information accuracy (IIA, %): share of questions on which the selected option reasons correctly with the implicit information in the story context, on the 2,246 MixRea four-option explicit-implicit reasoning questions over 673 story contexts from the Possible Stories dataset, read from which of the designed options the model selects; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 76.5 |
| 2 | Llama 3.1 405B | 71.1 |
| 3 | DeepSeek R1 | 70.9 |
| 4 | Claude 3.7 Sonnet | 70.1 |
| 5 | Llama 3.1 70B | 67.6 |
| 6 | Gemini 2.5 Flash | 67.1 |
| 7 | GPT-4o | 66.6 |
| 8 | Qwen 2.5 32B | 66.4 |
| 9 | DeepSeek V3 | 65.4 |
| 10 | Llama 3 70B | 65.4 |
| 11 | Gemma 3 27B | 65.2 |
| 12 | Gemma 2 27B | 65.1 |
| 13 | Qwen Max | 65 |
| 14 | Qwen 2 7B | 64.2 |
| 15 | Qwen 2.5 72B | 63.9 |
Interactive version: theaggregate.ai/benchmark?slug=mixrea-implicit-information-accuracy · How It Works · Data refreshed daily, snapshot 2026-10-07.