LLaVA-NeXT-110B: benchmark results
Provider: LLaVA Team. Access: Open.
Aggregate Rating 1522 ± 37, rank #1954 of 10552 rated models, from 27 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CoreCognition Image + Video Intentionality | 82.87 | Accuracy (%) | 97.5 |
| CoreCognition Image + Video Hierarchy | 65.94 | Accuracy (%) | 81.9 |
| CoreCognition Image + Video Mechanicalreasoning | 66.67 | Accuracy (%) | 81.5 |
| CoreCognition Image + Video Permanence | 45.1 | Accuracy (%) | 73.9 |
| LifeEval (Multiple-Choice) - Safety and Feasibility Assessment | 76.22 | Accuracy (%) on the 328 multiple-choice Safety and Feasibili | 72 |
| CoreCognition Image + Video Intuitivephysics | 57.27 | Accuracy (%) | 70.2 |
| LifeEval (Open-Ended) - Safety and Feasibility Assessment | 48.19 | GPT-5 judge score against the reference answer (0 to 1 in st | 60 |
| CoreCognition Image + Video Mean | 54.88 | Accuracy (%) | 56.3 |
| LifeEval (Multiple-Choice) | 65.45 | Accuracy (%) on LifeEval's 2,069 multiple-choice questions a | 56 |
| LifeEval (Multiple-Choice) - Contextual Knowledge Retrieval | 70.26 | Accuracy (%) on the 390 multiple-choice Contextual Knowledge | 56 |
| LifeEval (Multiple-Choice) - Dynamic Task Reasoning | 60.81 | Accuracy (%) on the 370 multiple-choice Dynamic Task Reasoni | 56 |
| LifeEval (Multiple-Choice) - Goal-oriented Planning | 59.95 | Accuracy (%) on the 377 multiple-choice Goal-oriented Planni | 56 |
Interactive version: theaggregate.ai/model?slug=llava-next-110b · How It Works · Data refreshed daily, snapshot 2026-11-16.