Llama 3.2 90B Instruct: benchmark results
Meta's largest open vision model: Llama 3.1 70B with a cross-attention image adapter, tuned for visual reasoning and captioning (September 2024). Provider: Meta. Released 2024-09-25. Access: Open.
Aggregate Rating 1554 ± 23, rank #601 of 1656 rated models, from 32 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BALROG BabyAI (VLM) | 66 | Progress (%) | 77.8 |
| LLM Stats (AI2D) | 92.3 | Score (%) | 75.8 |
| BALROG BabaIsAI (VLM) | 21.9 | Progress (%) | 66.7 |
| BALROG BabaIsAI (LLM) | 43.9 | Progress (%) | 60.5 |
| BALROG Official VLM Leaderboard | 21 | % Progress (percentage) | 55.6 |
| BALROG BabyAI (LLM) | 72 | Progress (%) | 53.9 |
| LLM Stats (ChartQA) | 85.5 | Score (%) | 52 |
| BALROG Crafter (LLM) | 31.68 | Progress (%) | 50 |
| BALROG Official LLM Leaderboard | 27.3 | % Progress (percentage) | 48.5 |
| SuperCLUE General (October 2024) - Science | 75.89 | Score | 46.3 |
| SuperCLUE General (October 2024) - Hard | 36.84 | Score | 39 |
| SuperCLUE General (October 2024) - Overall | 60.58 | Score | 36.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-2-90b-instruct · How It Works · Data refreshed daily, snapshot 2026-10-20.