Mistral Small 4 Pricing
Mistral · Live
Mistral Small 4 costs $0.15 per million input tokens and $0.6 per million output tokens on the Mistral API. Cached input reads are billed at $0.015 per million tokens. The model supports extended reasoning, image input, tool use. Released March 16, 2026. Pricing verified November 4, 2026.
Mistral Small 4 API pricing
Mistral Small 4 API pricing is $0.15 per million input tokens and $0.6 per million output tokens, verified November 4, 2026.
- Input (per 1M tokens)
- $0.15
- Output (per 1M tokens)
- $0.6
- Cached input (per 1M tokens)
- $0.01
- Batch discount
- 50%
- Released
- March 16, 2026
Capabilities
Mistral Small 4 supports extended reasoning, image input, tool use.
Price history
See full Mistral Small 4 price historyNo Mistral Small 4 price changes have been recorded yet.
No price snapshots recorded yet.
Frequently asked questions
Short answers to the most common questions about Mistral Small 4 pricing, limits, and capabilities.
Notes
Unified small-tier model released March 16, 2026 — combined what used to be three separate products (Magistral's reasoning, Pixtral's multimodal vision, Devstral's agentic coding) into one configurable model. 119B total parameters, 6B active per token. Single hybrid covering instruct, reasoning, and coding at budget rates. 50% batch API discount available. Open-weight under Apache 2.0. [2026-09-28] Verification attempt produced no clear result — recheck next run. [2026-10-11] Read directly from https://docs.mistral.ai/inference/pricing: $0.15/$0.60 confirmed. Cache Read empty → filled $0.015 per same page.