Llama 3.2 90B Vision Instruct — benchmark results

Meta Llama 3.2 90B vision instruction-tuned checkpoint. Provider: Meta. Released 2024-09-25. Access: Open.

Unified ELO 1478 ± 8, rank #879 of 1776 rated models, from 450 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
HELM SeaHELM - IndicQA57.23SQuAD macro-averaged F1 score100
OpenVLM MMT-Bench - Face Detection95Score (%)100
OpenVLM MMT-Bench - Writing Poetry From Image85Score (%)99.8
OpenVLM MMBench V1.1 EN - Function Reasoning97.8Accuracy (%)99.5
OpenVLM MMT-Bench - Film and Television Recognition100Score (%)99
OpenVLM MMT-Bench - Traffic Anomaly Detection70Score (%)99
OpenVLM MMT-Bench - Plant Recognition95Score (%)98.1
OpenVLM MMT-Bench - VR85.3Score (%)97.8
OpenVLM MMT-Bench - Abstract Visual Recognition90Score (%)96.8
OpenVLM MMT-Bench - Sign Language Recognition50Score (%)96.8
OpenVLM MMT-Bench - AR58Score (%)96.6
OpenVLM MMT-Bench - GUI General65Score (%)96.6

Interactive version: theaggregate.ai/model?slug=llama-3-2-90b-vision-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.