LLaVA-v1-7B: benchmark results

Provider: LLaVA Team. Released 2023-04-17.

Aggregate Rating 1231 ± 50, rank #9224 of 10489 rated models, from 399 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenVLM MMT-Bench - Font Recognition35Score (%)88.1
OpenVLM MMT-Bench - Deepfake Detection60Score (%)83
OpenVLM COCO Captions5.5CIDEr74.8
OpenVLM MMT-Bench - GUI Install45Score (%)71.6
OpenVLM MMMU - Energy and Power46.7Accuracy (%)71
OpenVLM MMT-Bench - Action Quality Assessment25Score (%)69.7
OpenVLM MMT-Bench - Vehicle Recognition100Score (%)69.4
OpenVLM MMMU - Biology46.7Accuracy (%)69.1
OpenVLM MMT-Bench - Handwritten Retrieval30Score (%)67.2
OpenVLM MMT-Bench - Image Quality Assessment45Score (%)66.5
OpenVLM MMT-Bench - Google Apps35Score (%)61.2
OpenVLM MMT-Bench - Chemical Apparatus Recognition60Score (%)57.8

Interactive version: theaggregate.ai/model?slug=llava-v1-7b · How It Works · Data refreshed daily, snapshot 2026-11-20.