Aria: benchmark results
Provider: Rhymes AI. Released 2024-10-10. Access: Open.
Aggregate Rating 1512 ± 14, rank #2167 of 10916 rated models, from 1025 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MEGA-Bench Task - MuMA Theory Of Mind Belief Of Goal | 80 | Task Score (%) | 100 |
| OpenVLM MMT-Bench - Temporal Sequence Understanding | 65 | Score (%) | 100 |
| OpenVLM MMT-Bench - Behavior Anomaly Detection | 80 | Score (%) | 99.3 |
| OpenVLM MMT-Bench - Animal Keypoint Detection | 55 | Score (%) | 99 |
| OpenVLM AI2D - Eclipses | 100 | Accuracy (%) | 98.6 |
| OpenVLM MMT-Bench - Multiple Image Captioning | 100 | Score (%) | 97.6 |
| LongVideoBench | 65 | Test Total Accuracy (%) | 97.1 |
| MLVU-Test - Anomaly Recognition | 64.1 | Multiple-choice accuracy (%) | 97.1 |
| OpenVLM MMT-Bench - Traffic Participants Understanding | 70 | Score (%) | 97.1 |
| OpenVLM MMT-Bench - Tech Engineering | 55.6 | Score (%) | 96.4 |
| MEGA-Bench Task - Reward Models I2T Reward | 71.4 | Task Score (%) | 96.2 |
| OpenVLM MMT-Bench - AUD | 67 | Score (%) | 96.1 |
Interactive version: theaggregate.ai/model?slug=aria · How It Works · Data refreshed daily, snapshot 2026-11-08.