Mistral AI · AI model
Ministral 3 14B
Dense 14B model, largest in the Ministral 3 family, released Dec 2, 2025 under Apache 2.0. Text and image input, 256k-token context, text output. $0.20 input / $0.20 output per 1M tokens.
Last verified Available 6 sources
Key specs
- Provider
- Mistral AI
- Family
- Ministral 3
- API identifier
- ministral-14b-2512
- Availability
- Available
- Release date
- 2025-12-02
- 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 Up to 8 images per the technical documentation
- 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 None API model lists text output only; a separate Reasoning variant is published as open weights
- Tool calling Yes Function Calling listed among the model's features
- Web search Unknown
"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 | $0.20 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 | $0.02 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 | $0.20 per 1M tokens | 2026-10-12 | 2026-10-25 | API pricing | Mistral Docs Standard mode list price |
| Batch input tokens | base | $0.10 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 | $0.10 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; no effort label; Stirrup harness (shell + web browsing) | 247 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
- #28 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 Ministral 3 14B
Guides
About Ministral 3 14B
Ministral 3 14B is the largest model in the Ministral 3 family of dense models (14B parameters) and is described by Mistral as optimized for local deployment. It accepts text and image input (up to 8 images) and has a 256k-token context window. Base, instruct and reasoning weights are released under the Apache 2.0 license; the API model (ministral-14b-2512) returns text output.
Strengths and weaknesses
Strengths
- Apache 2.0 open weights (base, instruct and reasoning variants)
- Text and image input with a 256k-token context window
- Supports function calling, structured outputs, Document QnA and batch processing
- List price of $0.20 input / $0.20 output per 1M tokens
Weaknesses
- No reasoning mode in the API (text output only)
- Agents & Conversations API and built-in tools are not listed among the model's supported features
- No audio or video input
Sources
Last verified Page updated 2026-10-26