LLaVA-Next-Yi-34B: benchmark results
Provider: LLaVA Team. Released 2024-01-30. Access: Open.
Aggregate Rating 1427 ± 40, rank #4824 of 10916 rated models, from 405 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMMU - Design | 90 | Accuracy (%) | 98.9 |
| OpenVLM MME - Color | 195 | Score | 98.7 |
| OpenVLM MMT-Bench - Jigsaw Puzzle Solving | 45 | Score (%) | 98.3 |
| OpenVLM MMT-Bench - Traffic Anomaly Detection | 65 | Score (%) | 97.8 |
| OpenVLM MMT-Bench - Polygon Localization | 70 | Score (%) | 96.1 |
| OpenVLM MMT-Bench - PLP | 53.3 | Score (%) | 95.9 |
| OpenVLM SEEDBench IMG - Visual Reasoning | 82.5 | Accuracy (%) | 95.8 |
| OpenVLM POPE | 89.6 | Overall (%) | 95.3 |
| OpenVLM MMT-Bench - Social Relation Recognition | 70 | Score (%) | 94.4 |
| OpenVLM MMT-Bench - Reasoning Segmentation | 73.7 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Pixel Localization | 45 | Score (%) | 93 |
| OpenVLM MME - Scene | 166.2 | Score | 92.9 |
Interactive version: theaggregate.ai/model?slug=llava-next-yi-34b · How It Works · Data refreshed daily, snapshot 2026-11-08.