Mistral-Large-3-675B-Instruct-2512 — benchmark results
Mistral's open Large 3 sparse MoE (675B total, 41B active), December 2025. Provider: Mistral. Released 2025-12-02. Access: Open.
Unified ELO 1618 ± 23, rank #439 of 1776 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RewardBench 2 Safety | 90.33 | Accuracy (%) | 84.3 |
| RewardBench 2 Focus | 90.51 | Accuracy (%) | 77.5 |
| Creative Writing v3 | 1400.2 | Elo score (self-reported) | 59.4 |
| RewardBench 2 Factuality | 69.32 | Accuracy (%) | 51 |
| Judgemark v2.1 | 68.2 | Judgemark Score (0-100) | 43.5 |
| MATH-MC Level 2 | 97.39 | Accuracy (%) | 36.8 |
| MATH-MC Level 5 | 93.94 | Accuracy (%) | 35.3 |
| EQ-Bench Longform Writing | 48.3 | Writing Score (0-100) | 35.2 |
| MATH-MC Level 1 | 96.98 | Accuracy (%) | 33.8 |
| MATH-MC Level 3 | 96.68 | Accuracy (%) | 33.8 |
| MATH-MC Level 4 | 95.75 | Accuracy (%) | 32.4 |
| GSM-MC | 97.8 | Accuracy (%) | 32.1 |
Interactive version: theaggregate.ai/model?slug=mistral-large-3-675b-instruct-2512 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.