| Llama 4 Maverick | Mistral Large 2 | |
|---|---|---|
| arenaElo | 1322 | 1315 |
| humanEval | 82 | 80.5 |
| sweBench | 45 | 38 |
| mmlu | 85.1 | 84 |
| gpqa | 54.3 | 52 |
| aime | 60 | 55 |
| Price in/out ($/M) | $0.2 / $0.8 | $2 / $6 |
Llama 4 Maverick has the bigger context (1M vs 128K) and stronger vision; Mistral Large 2 is EU-based with better European-language nuance and GDPR posture. Pick Llama for technical workloads, Mistral for EU enterprise.
Llama 4 Maverick · Mistral Large 2 · All AI model comparisons