Side-by-side comparison of Llama 3.3 70B (Meta) and Mistral Large (Mistral AI) — benchmarks, pricing, and capabilities.
| Llama 3.3 70B Meta | Mistral Large Mistral AI | |
|---|---|---|
| Category | LLMs | LLMs |
| Specifications | ||
| Context Window | 128K | 128K |
| Pricing (per 1M tokens) | ||
| Input Cost | — | $2.00 |
| Output Cost | — | $6.00 |
| Performance | ||
| Overall Score | 86.5 | 87.1 |
| ARC-Challenge | — | 94.0 |
| BigBench Hard | 81.0 | 80.1 |
| Chatbot Arena ELO | — | 1270.0 |
| DROP | 81.5 | 80.5 |
| GSM8K | 90.5 | 89.5 |
| HumanEval | 84.5 | 84.2 |
| MATH | 69.0 | 68.2 |
| MMLU | 83.8 | 72.2 |
| TruthfulQA | — | 67.2 |
| WinoGrande | — | 80.4 |
| Community | ||
| User Rating | ★ 4.4 | ★ 4.3 |
| Reviews | 310 | 312 |