API pricing modifiers: batch, cached input, regional inference
Catalog row in Vendor License Rules · Cited
- Kind
- Counting
- Statement
- Mistral pricing FAQ and API pricing (as of 2026-09-26; pages undated): input and output tokens are counted separately per million tokens; batch processing reduces the price by 50%; cached input tokens reduce input cost by up to 90% for repeated prompts; regional (global or EU) inference endpoints add 10% on supported models. OCR is priced per 1,000 pages, speech models per minute, and tool APIs per call.
- Applies when
- Pay-as-you-go API use.
- Applies to
- Mistral Studio API.
- Related metrics
- Related programs