All comparisons
HellaSwag 93.8 95.7 HumanEval 84.8 85.5 Massive Multitask Language Understanding 84.0 88.5 Chatbot Arena 1265.0 1302.0 BFCL 38.4 — MMLU PRO 69.7 75.9 LiveCodeBench — 49.6 Aider Polyglot — 48.4 BigCodeBench — 50.0
Mistral Large 2 vs DeepSeek V3
Side-by-side comparison of specifications, capabilities, and pricing.
| Released | 2025-11-25 | 2024-12-26 |
| Context window | 128K | 64k |
| Parameters | 123B | 671B |
| Architecture | decoder only | mixture of experts |
| License | True | Open Source |
| Knowledge cutoff | 2025-07 | 2024-04 |
Capabilities | ||
| Vision | ||
| Multimodal | ||
| Reasoning | ||
| Function calling | ||
| Tool use | ||
| Structured Outputs | ||
| Code execution | ||
Availability | ||
| Providers | ||
Benchmarks