Side-by-side comparison of Qwen 2.5 (Alibaba) and Mistral Large (Mistral AI) — benchmarks, pricing, and capabilities.
| Qwen 2.5 Alibaba | Mistral Large Mistral AI | |
|---|---|---|
| Category | LLMs | LLMs |
| Specifications | ||
| Context Window | 128K | 128K |
| Pricing (per 1M tokens) | ||
| Input Cost | Free | $2.00 |
| Output Cost | Free | $6.00 |
| Performance | ||
| Overall Score | 83.8 | 87.1 |
| ARC-Challenge | — | 91.1 |
| BigBench Hard | 75.8 | 80.1 |
| Chatbot Arena ELO | — | 1105.0 |
| DROP | 76.0 | 80.5 |
| GSM8K | 85.0 | 64.9 |
| HumanEval | 76.2 | 84.2 |
| MATH | 61.2 | 68.2 |
| MMLU | 81.5 | 83.5 |
| TruthfulQA | 57.0 | 77.4 |
| WinoGrande | — | 87.7 |
| Community | ||
| User Rating | ★ 4.3 | ★ 4.3 |
| Reviews | 198 | 312 |