Side-by-side comparison of Gemini 1.5 Pro (Google) and Codestral (Mistral AI) — benchmarks, pricing, and capabilities.
| Gemini 1.5 Pro Google | Codestral Mistral AI | |
|---|---|---|
| Category | LLMs | Code Models |
| Specifications | ||
| Context Window | 1.0M | 32K |
| Pricing (per 1M tokens) | ||
| Input Cost | $1.25 | $0.20 |
| Output Cost | $5.00 | $0.60 |
| Performance | ||
| Overall Score | 90.8 | 86.2 |
| ARC-Challenge | 92.2 | — |
| BigBench Hard | 85.1 | — |
| Chatbot Arena ELO | 1240.0 | — |
| DROP | 85.8 | — |
| GSM8K | 93.2 | 84.0 |
| HumanEval | 89.1 | 92.5 |
| MATH | 74.0 | 60.5 |
| MBPP | — | 88.0 |
| MMLU | 86.4 | 79.0 |
| TruthfulQA | 69.8 | — |
| WinoGrande | 80.1 | — |
| Community | ||
| User Rating | ★ 4.5 | ★ 4.6 |
| Reviews | 654 | 289 |