Unison: leaderboard
Unison (Fudan, 2026) evaluates unified multimodal models on 2,169 tasks that need understanding and image generation together rather than separately. The overall score is the mean of four unified dimension scores (internal consistency, understanding-guided generation, generation-guided understanding and mutual enhancement), scored by an 8B Unison-Judge and task metrics. Higher is better.
Metric: Overall score (%): mean of the four unified dimension scores (internal consistency, understanding-guided generation, generation-guided understanding, mutual enhancement) on Unison's 2,169 samples, scored by the authors' Unison-Judge and task metrics; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 71.3 | #176 |
| 2 | Gemini 3 Pro | 69.8 | #137 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=unison · How It Works · Data refreshed daily, snapshot 2026-11-13.