Taichu-VLR-3B — benchmark results
Provider: Other. Released 2025-03-22. Access: Open.
Unified ELO 1477 ± 12, rank #932 of 1841 rated models, from 98 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LMM Reasoning - WeMath - InadequateGeneralization (Loose) | 15.4 | Accuracy (%) | 93.8 |
| OpenVLM MathVista - FQA | 73.6 | Accuracy (%) | 88.1 |
| Open LMM Reasoning - WeMath - Angles and Length | 53.7 | Accuracy (%) | 87.7 |
| OpenVLM MathVista - STA | 82.1 | Accuracy (%) | 87.1 |
| OpenVLM MathVista - LOG | 29.7 | Accuracy (%) | 81.3 |
| Open LMM Reasoning - LogicVista - inductive | 32.7 | Accuracy (%) | 80.2 |
| Open LMM Reasoning - WeMath - Correspondence of Coordinates and Positions | 80 | Accuracy (%) | 79.6 |
| Open LMM Reasoning - MathVision - topology | 26.1 | Accuracy (%) | 76.5 |
| OpenVLM MathVista - SCI | 66.4 | Accuracy (%) | 74.5 |
| Open LMM Reasoning - MathVision - transformation geometry | 27.4 | Accuracy (%) | 74 |
| Open LMM Reasoning - LogicVista - spatial | 28.2 | Accuracy (%) | 71.6 |
| Open LMM Reasoning - MathVerse - Expression | 18.8 | Accuracy (%) | 71.4 |
Interactive version: theaggregate.ai/model?slug=taichu-vlr-3b · How It Works · Data refreshed daily, snapshot 2026-07-25.