MCBench: leaderboard
MCBench (2026) asks omni-modal models whether a situation is safe or unsafe, where the answer depends on combining an image, an audio clip and a spoken-style utterance; 1,196 scenarios come in minimally differing safe and unsafe pairs across physical, social, illegal and property harm. The board is the average accuracy in percent over the safe and unsafe cases of the four categories (random 50); higher is better.
Metric: Average accuracy (%) over the eight category-by-label cells (safe and unsafe scenarios of physical, social, illegal and property harm) of safe-or-unsafe judgments from an image, an audio clip and an utterance, chain-of-thought prompt, GPT-4o judge maps free-form answers to a label, mean of five runs; random 50; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 64.4 | #453 |
| 2 | Gemma 4 12B (IT) | 61.9 | #693 |
| 3 | Qwen2.5-Omni-7B | 55.2 | #1789 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=mcbench · How It Works · Data refreshed daily, snapshot 2026-11-13.