VLM-R1-3B-Math-0305 — benchmark results
Provider: Other. Released 2025-03-05. Access: Open.
Unified ELO 1486 ± 10, rank #898 of 1841 rated models, from 98 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LMM Reasoning - LogicVista - spatial | 33.3 | Accuracy (%) | 93.8 |
| OpenVLM MathVista - LOG | 35.1 | Accuracy (%) | 90.3 |
| Open LMM Reasoning - WeMath - Correspondence of Coordinates and Positions | 83.3 | Accuracy (%) | 90.1 |
| Open LMM Reasoning - MathVerse - Analytic | 46.5 | Accuracy (%) | 85.7 |
| OpenVLM MathVista - FQA | 72.1 | Accuracy (%) | 85.4 |
| OpenVLM MathVista - VQA | 57.5 | Accuracy (%) | 84.7 |
| Open LMM Reasoning - MathVision - descriptive geometry | 25 | Accuracy (%) | 84.2 |
| OpenVLM MathVista - STA | 79.1 | Accuracy (%) | 78.7 |
| Open LMM Reasoning - WeMath - Angles and Length | 48.4 | Accuracy (%) | 75.3 |
| OpenVLM MathVista - SCI | 65.6 | Accuracy (%) | 71.6 |
| Open LMM Reasoning - MathVision - transformation geometry | 26.2 | Accuracy (%) | 69.4 |
| Open LMM Reasoning - WeMath - InadequateGeneralization (Loose) | 13.1 | Accuracy (%) | 67.3 |
Interactive version: theaggregate.ai/model?slug=vlm-r1-3b-math-0305 · How It Works · Data refreshed daily, snapshot 2026-07-25.