llava-v1.6-34B-hf: benchmark results
Provider: LLaVA Team. Access: Open.
Aggregate Rating 1443 ± 41, rank #4096 of 10916 rated models, from 34 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DrawingVQA - Domain-Expert Reasoning | 44.4 | Accuracy (%; reasoning depth R3, 27 domain-expert reasoning | 75 |
| MERA Multi - ruTiE-Image | 48.7 | Score (%) | 72.4 |
| MERA Multi - LabTabVQA | 11.4 | Score (%) | 66.1 |
| MERA Multi - ruHHH-Image | 14.4 | Score (%) | 62.7 |
| MERA Multi - ruNaturalScienceVQA | 20.7 | Score (%) | 61.9 |
| MERA Multi - WEIRD | 31.4 | Score (%) | 61 |
| MERA Multi - Image | 19 | Score (%) | 58.5 |
| VRIQ (Abstract) - Sequence Completion | 25 | Accuracy (%) on the 200 abstract-domain VRIQ sequence comple | 58.3 |
| MERA Multi - SchoolScienceVQA | 25.8 | Score (%) | 55.9 |
| MERA Multi - ruCommonVQA | 30.1 | Score (%) | 50.8 |
| VRIQ (Abstract) - Matrix Prediction | 26 | Accuracy (%) on the 200 abstract-domain VRIQ matrix predicti | 50 |
| VRIQ (Natural) - Matrix Prediction | 43 | Accuracy (%) on the 100 natural-domain VRIQ matrix predictio | 45.8 |
Interactive version: theaggregate.ai/model?slug=llava-v1-6-34b-hf · How It Works · Data refreshed daily, snapshot 2026-11-08.