Model overview
Positioned between the small and medium tiers, mistral-large-latest asks 0.5 per million input tokens and 1.5 for output, a three-to-one ratio gentler than the five-to-one of the medium tier, which charges 1.5 and 7.5, three times more to read and five times more to write. Its declared capabilities are multimodal and multilingual only, without the coding, reasoning or agentic markers found elsewhere in the range, so it reads as a generalist rather than a specialist. Code work is better served by codestral-latest, which sits 40 percent below it on both sides at 0.3 and 0.9. Cached input costs 0.05, one tenth of the standard rate.
Billing conditions: Standard API price. Mistral publishes a 90% discount for cached input; cache-write prices are not published separately.
- Context window
- 0 tokens
- Knowledge cutoff
- —
- Price last verified
- First observed
Descriptions, capabilities, and tariffs come from the linked source. First-party prices take priority; OpenRouter fallbacks are labelled.