Mistral Medium 3.1
Neural Network
Mistral Medium 3.1 is an enterprise language model by Mistral AI, available via API in BotHub for professional text tasks.
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Mistral Medium 3.1?
Mistral Medium 3.1 is an updated version of Mistral Medium 3 by Mistral AI: an enterprise-grade language model designed to provide advanced system capabilities at significantly lower operating costs. It focuses on balancing quality and cost, while the batch variant is designed for bulk processing when high request volume is needed but instant responses are not. Through BotHub, the model is accessible without a VPN or foreign card; payment is made in rubles per token, and tokens do not expire. Over 250 models are available in the same window, allowing you to compare responses to a single prompt and choose the best one. A unified OpenAI-compatible API connects Mistral Medium 3.1 to your services and allows you to swap it for another model without rewriting the integration. Companies have access to contracts, invoices, electronic document management (EDO), and an admin panel with employee limits. It is useful for batch processing arrays of documents and requests, structuring raw text, preparing drafts, editing materials, and building internal assistants on top of your knowledge base.Frequently asked questions about Mistral Medium 3.1
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.
This is a Mistral text model: it writes and edits text, analyzes code, answers based on uploaded documents and images, and calls external functions. The batch variant is convenient when you need to run a large volume of similar requests — classification, labeling, summarization, and data extraction from files.
The single response limit is 104,857 tokens, and the context window holds 131,072 tokens. This is enough for a voluminous document: a detailed instruction, a chapter translation, a large report, or a detailed code analysis. Keep in mind that the window includes both your request and the response.
A separate reasoning mode is not noted in our data for this model, so it answers immediately. If the task requires step-by-step logic, ask it in the prompt to break down the solution into points — this usually significantly improves results for math and complex analysis.
Yes, both mechanisms are supported. The model can call functions and tools, and return responses in a strict JSON structure according to your schema — this is convenient for agents, parsers, and integrations where the result goes directly into code without manual text parsing.
Images and documents — yes: you can send a screenshot, diagram, photo, or file and ask it to describe, extract data, or verify a table. Audio and video are not noted in our data. The model outputs text; for audio and video in BotHub, switch to a specialized model.
There is no precise data on languages in our model card, so we cannot provide guarantees. The best way is a short test on your material: submit a typical request and evaluate the response. In BotHub, payment is per token, tokens do not expire, and you can compare it with other models in the same window.