Side-by-side comparison of DeepSeek Coder V2 (DeepSeek) and Codestral (Mistral AI) — benchmarks, pricing, and capabilities.
| DeepSeek Coder V2 DeepSeek | Codestral Mistral AI | |
|---|---|---|
| Category | Code Models | Code Models |
| Specifications | ||
| Context Window | 128K | 32K |
| Pricing (per 1M tokens) | ||
| Input Cost | $0.14 | $0.20 |
| Output Cost | $0.28 | $0.60 |
| Performance | ||
| Overall Score | 84.5 | 86.2 |
| ARC-Challenge | 89.7 | — |
| GSM8K | 83.2 | 84.0 |
| HumanEval | 90.8 | 92.5 |
| MATH | 59.8 | 60.5 |
| MBPP | 86.5 | 88.0 |
| MMLU | 78.2 | 79.0 |
| TruthfulQA | 72.7 | — |
| WinoGrande | 85.9 | — |
| Community | ||
| User Rating | ★ 4.5 | ★ 4.6 |
| Reviews | 445 | 289 |