Llama 3.2 11B Instruct — benchmark results

Meta Llama 3.2 11B vision instruction-tuned checkpoint. Provider: Meta. Released 2024-09-25. Access: Open.

Unified ELO 1394 ± 7, rank #1253 of 1776 rated models, from 639 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AGC-Bench - schnovel1.91Dataset z-score98.8
Open LMM Reasoning - WeMath - InsufficientKnowledge (Loose)80Accuracy (%)97.5
Open LMM Reasoning - WeMath - RoteMemorization (Loose)36Accuracy (%)97.5
OpenVLM MMT-Bench - Abstract Visual Recognition90Score (%)96.8
AGC-Bench - puntuguese1.02Dataset z-score96.3
OpenVLM MMT-Bench - National Flag Recognition100Score (%)94.9
OpenVLM MMT-Bench - Table Structure Recognition100Score (%)94.7
OpenVLM MMT-Bench - Logo and Brand Recognition100Score (%)94.4
AGC-Bench - creatset1.47Dataset z-score93.9
OpenVLM MMT-Bench - Humanities and Social Science72.7Score (%)93.9
OpenVLM MMT-Bench - Astronomical Recognition88.9Score (%)93
AGC-Bench - slang_generation1.1Dataset z-score92.7

Interactive version: theaggregate.ai/model?slug=llama-3-2-11b-instruct · How the rankings work · Data refreshed daily, snapshot 2026-07-22.