Ministral 3B 2512
Neural Network
Ministral 3B 2512 is a compact Mistral AI model with image support: Ministral 3B API for fast text and visual tasks.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 131 072 tokens
- Max answer length
- 104 857 tokens
- Capabilities
- function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Ministral 3B 2512?
Ministral 3B 2512 is the most compact model in the Ministral 3 family by Mistral AI: a small language model that, despite its modest size, works with both text and images. It is chosen for its response speed and cost-effectiveness—ideal for high-volume requests that do not require heavy reasoning. Through BotHub, you can access it from Russia without a VPN or foreign card, pay with a Russian card in rubles based on actual usage (tokens consumed), and unused balances do not expire. Over 250 models are available in one window: you can switch and compare answers, and a unified OpenAI-compatible API allows you to change models in your product without rewriting the integration. Traffic is encrypted via AES-GCM, chat content is not saved, and for companies, we offer contracts, invoices, electronic document management (EDO), and an admin panel with limits. It is useful for mass processing of short requests, text classification, maintaining a fast-response chat assistant, drafting formulations, rewriting fragments, analyzing image and screenshot content, or building a prototype before moving to a larger model.Frequently asked questions about Ministral 3B 2512
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
It is a compact Mistral text model with a 131,072 token context window. Suitable for daily work with text and documents: summarization, rewriting, answering based on uploaded files, code assistance, extracting data from images, and embedding into applications via the BotHub API.
The output limit is 104,857 tokens per response, which is enough for a large document, detailed analysis, or a large code fragment in its entirety. The 131,072 token context allows you to keep source materials in the dialogue and receive long results without breaking them into parts.
A separate chain-of-thought mode is not noted in our data. However, you can ask the model to break down the task step-by-step directly in the prompt—this usually improves quality for logic and calculations. If you need strong reasoning, switch to a reasoning model in the same window.
Yes, the model supports function calling and structured JSON output. You describe the schema and tools, and receive a predictable response for parsing and connecting external services. In BotHub, this works via a unified OpenAI-compatible API, so you won't have to rewrite your integration.
Images and documents are accepted as input: you can ask it to describe a picture, recognize a diagram or screenshot, analyze a PDF, and answer based on its content. Audio and video processing is not noted in our data—for such tasks, choose a specialized model from the catalog.
Mistral models were created as multilingual, so the model works with Russian-language queries. The exact quality depends on the task, so test it on your own data. In BotHub, you only pay for used tokens and can compare several models in one window to choose the best one.