Llama 3.2 Vision 11B Instruct: benchmark results

Provider: Meta. Released 2024-09-25. Access: Open.

Unified ELO 1414 ± 35, rank #2019 of 2928 rated models, from 13 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
BenchTable - Utility56.6Weighted Score (%)56.5
SnakeBench18TrueSkill Rating31.5
Wolfram LLM Benchmarking Project23.9Correct Functionality (%)21.8
BenchTable - Tech19.7Weighted Score (%)21.5
BenchTable18.6Total Score (%)14.4
BenchmarkList ECI75.25Capability Index (ECI)8.3
Chatbot Arena (Vision - English)1016Arena Score7.3
Chatbot Arena (Vision)993Arena Score5.3
BenchTable - Reasoning-1.6Weighted Score (%)3.1
BenchTable - STEM1.9Weighted Score (%)2.9
SEAL - VISTA20.47Score1.6
BLUEX v24.18Score (self-reported)0

Interactive version: theaggregate.ai/model?slug=llama-3-2-vision-11b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-23.