MM-JudgeBias - Text Insertion: leaderboard
MM-JudgeBias (2026) tests multimodal LLM judges for compositional bias: the judge scores image, query and response triplets from 1 to 10 before and after a controlled perturbation. Bias-Conformity from 0 to 1, the stability of the score when query keywords are overlaid on the image; higher is better.
Metric: Bias-Conformity (0 to 1) on MM-JudgeBias (about 200 image-query-response triplets per bias type, 1,804 in all, each judged on a 1-10 scale before and after a controlled perturbation; mean of three runs): mean score invariance (1 minus the absolute score change over its largest possible value) when query keywords or the query text are overlaid on the image; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 26 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | LLaVA-Critic-72B | 0.942 | ||
| 2 | Gemini 3 Pro | High | 0.933 | #145 |
| 3 | Gemini 2.5 Pro | 0.93 | #276 | |
| 4 | GPT-5 Mini | High | 0.927 | #364 |
| 5 | Gemini 2.5 Flash | 0.918 | #463 | |
| 6 | O4 Mini | High | 0.909 | #368 |
| 7 | GPT-5.1 | High | 0.908 | #240 |
| 8 | GPT-4.1 Mini | 0.904 | #789 | |
| 9 | Claude Opus 4.5 | Thinking | 0.903 | #160 |
| 10 | Qwen 3 VL 8B (Thinking) | 0.898 | #910 | |
| 11 | Claude Sonnet 4.5 | Thinking | 0.893 | #261 |
| 12 | O3 | High | 0.884 | #237 |
| 13 | Qwen 3 VL 30B A3B Instruct | 0.883 | #895 | |
| 14 | Qwen 3 VL 8B Instruct | 0.88 | #1029 | |
| 15 | Gemini 2.0 Flash Lite | 0.878 | #1050 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mm-judgebias-text-insertion · How It Works · Data refreshed daily, snapshot 2026-11-21.