Side-by-side comparison of Codestral (Mistral AI) and Claude Computer Use (Anthropic) — benchmarks, pricing, and capabilities.
| Codestral Mistral AI | Claude Computer Use Anthropic | |
|---|---|---|
| Category | Code Models | AI Agents |
| Specifications | ||
| Context Window | 32K | 200K |
| Pricing (per 1M tokens) | ||
| Input Cost | $0.20 | $3.00 |
| Output Cost | $0.60 | $15.00 |
| Performance | ||
| Overall Score | 86.2 | 86.0 |
| GSM8K | 84.0 | — |
| HumanEval | 92.5 | — |
| MATH | 60.5 | — |
| MBPP | 88.0 | — |
| MMLU | 79.0 | — |
| SWE-bench Verified | — | 49.0 |
| Community | ||
| User Rating | ★ 4.6 | ★ 4.3 |
| Reviews | 289 | 210 |