LLaVA-LLaMA-3-8B: benchmark results
Provider: LLaVA Team. Released 2024-04-22. Access: Open.
Aggregate Rating 1330 ± 37, rank #8170 of 10916 rated models, from 405 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - MeViS | 65 | Score (%) | 100 |
| OpenVLM MMT-Bench - Pixel Recognition | 75 | Score (%) | 94.7 |
| OpenVLM MMT-Bench - 3D Indoor Recognition | 45 | Score (%) | 91.5 |
| OpenVLM MMT-Bench - MIA | 72.5 | Score (%) | 91.3 |
| OpenVLM MMT-Bench - Geometrical Perspective | 83.3 | Score (%) | 90.5 |
| OpenVLM MMT-Bench - Micro Expression Recognition | 40 | Score (%) | 89.8 |
| OpenVLM MME - Existence | 200 | Score | 89.7 |
| OpenVLM MMT-Bench - LVLM Response Judgement | 60 | Score (%) | 88.6 |
| OpenVLM MMT-Bench - Spot The Diff | 80 | Score (%) | 88.6 |
| OpenVLM MMMU - Architecture and Engineering | 50 | Accuracy (%) | 88.3 |
| OpenVLM MMT-Bench - Scene Recognition | 80 | Score (%) | 88.1 |
| OpenVLM MMT-Bench - One Shot Detection | 80 | Score (%) | 87.1 |
Interactive version: theaggregate.ai/model?slug=llava-llama-3-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.