OmniEarth - Change Mask Segmentation: leaderboard

Metric: Mean intersection over union (%) of predicted masks, Change Mask Segmentation (CMS) task of OmniEarth, zero-shot with each model's default inference settings; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 16 models tracked.

Top models

#ModelScoreOverall rank
1GLM-4.6V10.9#309
2GPT-4o10.7#333
3InternVL3-8B7.2#606
4Qwen 2.5 VL 72B Instruct6.7#364
5Gemini 2.0 Flash1.9#331

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=omniearth-change-mask-segmentation · How It Works · Data refreshed daily, snapshot 2026-10-11.