Side-by-side comparison of Codestral (Mistral AI) and Gemini 1.5 Pro (Google) — benchmarks, pricing, and capabilities.
| Codestral Mistral AI | Gemini 1.5 Pro Google | |
|---|---|---|
| Category | Code Models | LLMs |
| Specifications | ||
| Context Window | 32K | 1.0M |
| Pricing (per 1M tokens) | ||
| Input Cost | $0.20 | $1.25 |
| Output Cost | $0.60 | $5.00 |
| Performance | ||
| Overall Score | 86.2 | 90.8 |
| ARC-Challenge | — | 89.5 |
| BigBench Hard | — | 85.1 |
| Chatbot Arena ELO | — | 1126.0 |
| DROP | — | 85.8 |
| GSM8K | 84.0 | 79.4 |
| HumanEval | 92.5 | 89.1 |
| MATH | 60.5 | 74.0 |
| MBPP | 88.0 | — |
| MMLU | 79.0 | 85.3 |
| TruthfulQA | — | 69.8 |
| WinoGrande | — | 84.2 |
| Community | ||
| User Rating | ★ 4.6 | ★ 4.5 |
| Reviews | 289 | 654 |