Cambrian-S Release Image - MathVista M: leaderboard
Cambrian-S release paper; final Stage4 checkpoint. Benchmark-specific sampling of Section3 retained, dagger-marked video scores use paper protocol; no pretrained, intermediate-stage or LFP checkpoint merge. Appendix image cohort; only finalStage4 releases selected, intermediate Stage1/2/3 remain distinct source evidence. Metric: Accuracy (%). Higher is better. Source: https://arxiv.org/html/2511.04670v1
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Cambrian-S-7B | 50.6 | |
| 2 | Cambrian-1-8B | 49 | |
| 3 | Cambrian-S-3B | 43 | |
| 4 | Cambrian-S-1.5B | 39.7 | |
| 5 | Mini-Gemini-HD-8B | 37 | |
| 6 | LLaVA-NeXT-8B | 36.3 | |
| 7 | Cambrian-S-0.5B | 28.6 |
Interactive version: theaggregate.ai/benchmark?slug=cambrian-s-release-image-mathvista-m · How It Works · Data refreshed daily, snapshot 2026-10-17.