pixtral-large-2411: benchmark results
Mistral's Pixtral Large, a 124B multimodal model (123B decoder plus 1B vision encoder) built on Mistral Large 2, with a 128K context window supporting document, chart and image understanding. Provider: Mistral. Released 2024-11-18. Access: Open.
Aggregate Rating 1588 ± 20, rank #481 of 1661 rated models, from 101 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AGC-Bench - proparalogy | 0.87 | Dataset z-score | 95.1 |
| AGC-Bench - simile_generation | 0.92 | Dataset z-score | 87.5 |
| TextClass Benchmark | 1697.33 | Meta-Elo (imported) | 86 |
| AGC-Bench - analobench | 0.77 | Dataset z-score | 85.2 |
| AGC-Bench - creatset | 1.18 | Dataset z-score | 82.9 |
| AGC-Bench - permpst | 0.81 | Dataset z-score | 81.7 |
| AGC-Bench - metaphor_generation | 0.69 | Dataset z-score | 78 |
| AGC-Bench - c3_crosstalk | 0.81 | Dataset z-score | 76.8 |
| AGC-Bench - grapheval_review_advisor | 0.61 | Dataset z-score | 75.6 |
| AMVICC - MMVP Individual Accuracy - State and Condition | 83.33 | Individual accuracy (%; the 54 MMVP questions on 27 image pa | 75 |
| AGC-Bench - science_analogies | 0.49 | Dataset z-score | 73.8 |
| AGC-Bench - outline_to_story | 0.58 | Dataset z-score | 72.8 |
Interactive version: theaggregate.ai/model?slug=pixtral-large-2411 · How It Works · Data refreshed daily, snapshot 2026-10-23.