Ta sekcja nie została jeszcze przetłumaczona. Poniższa treść jest wyświetlana po angielsku, w wersji opublikowanej. Przeczytaj wersję angielską.
Mistral · Text model · GA
Mistral: a tiny and efficient model with text and vision docs.mistral.ai (opens in a new tab)
Compare with similar models Open the Model Finder
Figures checked 1 Nov 2026. A dash means not published on the sources we checked, so it stays blank rather than estimated.
US$ per 1M tokens, standard tier, short prompts. Output left blank where not published.
| Model | Input US$/1M | Output US$/1M |
|---|---|---|
| Ministral 3 3B | 0.1 | 0.1 |
| Step 3.5 Flash | 0.1 | 0.3 |
| Ministral 3 8B | 0.15 | 0.15 |
| Gemini 2.5 Flash-Lite | 0.1 | 0.4 |
| Ministral 3 14B | 0.2 | 0.2 |
| Lab | Mistral |
|---|---|
| Type | Text |
| Status | GA docs.mistral.ai (opens in a new tab) |
| Released | – |
| API access | Yes docs.mistral.ai (opens in a new tab) |
| API model ID | ministral-3b-2512 docs.mistral.ai (opens in a new tab) |
| API rate limits | – |
| Consumer app plans | – |
| Price at each level | – |
| Input per 1M tokens | US$0.1 docs.mistral.ai (opens in a new tab) |
| Output per 1M tokens | US$0.1 docs.mistral.ai (opens in a new tab) |
| Cached input | – |
| Batch in / out | – |
| Other pricing | – |
| Context window | – |
| Max output | – |
| Input types | text, image docs.mistral.ai (opens in a new tab) |
| Output types | text |
| Weights | Open docs.mistral.ai (opens in a new tab)Apache 2.0 |
| Reasoning | – |
| Tool use | – |
| Artificial Analysis index | – |
| Output speed | – |
| Time to first token | – |
| Reasoning levels | – |
| Lab benchmark claims | – |
| Limitations | – |
Ministral 3 3B has no published tool use, so we do not match it to MCP servers. Browse AI tools
Same type, closest on quality index and blended price. A similarity, not a ranking.
Compare Ministral 3 3B with Step 3.5 Flash and Ministral 3 8B
All 109 AI models · AI agents · How we check sources
Mistral 
Four Voxtral 3 models cover transcription in 50+ languages and text to speech, with a managed voice agent pipeline. Mistral offers EU, US and global hosted endpoints or self-hosting, so audio can stay on your own infrastructure. Transcription starts at $0.0035 per minute.
Source: Mistral (opens in a new tab)
Mistral
Mistral previewed Large 4 on 6 October and has since released the weights. It is a mixture-of-experts model with 52 billion active parameters and a 1M-token context, priced at $0.70 in and $2.10 out per 1M tokens. It is also on Microsoft Foundry and NVIDIA-hosted endpoints.
Source: Mistral (opens in a new tab)
