Mistral AI · AI model
Mistral Small 4
Hybrid MoE model (119B total, 6.5B active parameters) unifying instruct, reasoning and coding. Text and image input, 256k-token context, reasoning_effort none/high. Apache 2.0 open weights; $0.15/$0.60 per 1M tokens.
Last verified Available 8 sources
Key specs
- Provider
- Mistral AI
- Family
- Mistral Small
- API identifier
- mistral-small-2603
- Availability
- Available
- Release date
- 2026-03-16
- License
- Open source (Apache 2.0)
- Context window
- 256,000 tokens
- Max output
- 256,000 tokens
- Knowledge cutoff
- July 2025
- API access
- Yes
- Consumer app
- Unknown
Capabilities
Inputs
- Text input Yes
- Image input Yes Technical documentation lists image input maximum size as 256k tokens
- Audio input No Technical documentation marks audio input as not supported (N/A)
- Video input No Technical documentation marks video input as not supported (N/A)
- File input Yes PDF documents via Document QnA (OCR + model) in Chat Completions/Conversations; Document QnA is listed on the model page; uploaded files max 50 MB and 1,000 pages
Outputs
- Text output Yes
- Image output No Only text output listed; image generation exists separately as a built-in agent tool
- Audio output No Only text output listed
Features
- Reasoning Optional Adjustable via reasoning_effort parameter ('none' or 'high'); documented for the alias mistral-small-latest
- Tool calling Yes Function Calling listed for /v1/chat/completions and /v1/conversations
- Web search Yes Built-in web_search / web_search_premium tools via the Conversations and Agents APIs (not Chat Completions); model page lists Built-In Tools
"Unknown" means we have not verified this capability yet; it does not mean the model lacks it.
Pricing
Estimate your monthly cost →| Dimension | Tier | Price | Since | Verified | Source |
|---|---|---|---|---|---|
| Input tokens | base | $0.15 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Standard mode list price (Priority mode +75% and regional inference +10% not included) |
| Cached input tokens | base | $0.015 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Cached input price listed on the docs pricing page (10% of the input price) |
| Output tokens | base | $0.60 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Standard mode list price |
| Batch input tokens | base | $0.075 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Batch API: 50% of standard price (docs pricing page Batch mode, factor 0.5: 'Batching ... at a 50% discount'; also stated on mistral.ai/pricing); computed from the standard list price |
| Batch output tokens | base | $0.30 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Batch API: 50% of standard price (docs pricing page Batch mode, factor 0.5); computed from the standard list price |
Prices are shown in the currency the provider publishes; we never convert currencies.
Benchmarks
| Benchmark | Setting | Score | Reported by | Measured | Verified | Source |
|---|---|---|---|---|---|---|
| GDPval-AA v2.1 Elo, higher is better | v2.1, 'Mistral Small 4 (Reasoning)'; reasoning on, effort level not stated; Stirrup harness (shell + web browsing) | 362 Elo | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
Elo is a relative rating: values change as models are added and the scale is re-anchored, so always store the date a rating was read. For example, Anthropic's launch page gives 1846 for Claude Opus 5.5, while the leaderboard on 2026-10-12 shows 1866 (Max, Default Fallback). Earlier versions (Google cites "GDPVal-AA v2") are not comparable. Judging is done by an LLM, not by human experts. Anthropic models are listed with "Default Fallback", meaning refused steps are routed to a fallback model. AA-Briefcase and GDP.pdf are separate evaluations.
One row per result, without rank numbers: each result was measured under its own setting (effort level, harness, tools, test version), as stated by the source, and scores compare only within the same setting. Self-reported results come from the provider and have not been reproduced independently.
Rankings
- #26 of 39 in Best AI for business as of 2026-10-26
Derived from our published ranking snapshots; each ranking explains its criteria and weights.
Compare Mistral Small 4
Guides
- AI token pricing explained: input, output, cached and batch rates
- Reasoning models explained: thinking modes, effort and cost
- Multimodal AI explained: image, audio, video and file inputs, and what they cost
- Context windows explained: limits, long-context pricing and recall
- How to choose an AI model: a practical decision framework
About Mistral Small 4
Mistral Small 4 is a hybrid Mixture-of-Experts model that combines the instruct, reasoning (Magistral) and agentic coding (Devstral) lines in one model; the docs list 119B total and 6.5B active parameters. It accepts text and image input and has a 256k-token context window. Reasoning is controlled per request with reasoning_effort ('none' or 'high'). Weights are released under the Apache 2.0 license.
Strengths and weaknesses
Strengths
- Apache 2.0 open weights
- Text and image input with switchable reasoning (
reasoning_effort) - Lowest list price among the reasoning-capable models in this catalog ($0.15 input / $0.60 output per 1M tokens)
- Supports function calling, structured outputs, predicted outputs, batch processing and built-in tools
Weaknesses
- 256k-token context window, smaller than Mistral Large 4 (1M)
- No audio or video input
Sources
- Mistral Small 4 - Mistral AI | Mistral Docs
- Mistral Small 4 - Technical Documentation (for downstream providers)
- Mistral Small 4 - Public Summary of Training Content
- Document QnA | Mistral Docs
- Reasoning | Mistral Docs
- Websearch | Mistral Docs
- API pricing | Mistral Docs
- GDPval-AA v2.1 Leaderboard (Artificial Analysis)
Last verified Page updated 2026-10-26