LLaVA-NeXT-Video-DPO (34B): benchmark results
Provider: LLaVA Team.
Aggregate Rating 1489 ± 52, rank #2485 of 10556 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| VideoHallucer | 32.3 | Overall pair accuracy over five hallucination settings (%) | 73.3 |
| VideoOdyssey-V - Temporal Ordering | 36 | Multiple-choice accuracy (%) on the 89 temporal ordering que | 55.9 |
| VideoOdyssey-V - Attribute Recognition | 34 | Multiple-choice accuracy (%) on the 106 attribute recognitio | 38.2 |
| VideoOdyssey-V - Action Recognition | 26.6 | Multiple-choice accuracy (%) on the 192 action recognition q | 29.4 |
| VideoOdyssey-V - Captioning | 34.7 | Multiple-choice accuracy (%) on the 49 captioning questions | 29.4 |
| VideoOdyssey-V - Intentional Reasoning | 33.3 | Multiple-choice accuracy (%) on the 66 intentional reasoning | 29.4 |
| VideoOdyssey-V - Summarization | 30.8 | Multiple-choice accuracy (%) on the 65 summarization questio | 23.5 |
| VideoOdyssey-V - Emotional Reasoning | 31 | Multiple-choice accuracy (%) on the 71 emotional reasoning q | 20.6 |
| VideoOdyssey-V - OCR | 27.5 | Multiple-choice accuracy (%) on the 167 OCR questions (recog | 17.6 |
| VideoOdyssey-V - Spatial Reasoning | 24.6 | Multiple-choice accuracy (%) on the 61 spatial reasoning que | 17.6 |
| VideoOdyssey-V - Temporal Grounding | 26.2 | Multiple-choice accuracy (%) on the 65 temporal grounding qu | 17.6 |
| VideoOdyssey-V | 25.7 | Multiple-choice accuracy (%) over all questions of VideoOdys | 14.7 |
Interactive version: theaggregate.ai/model?slug=llava-next-video-dpo-34b · How It Works · Data refreshed daily, snapshot 2026-11-17.