Mistral Large 2
Neural Network
Mistral Large 2 is Mistral AI's flagship model for reasoning, code, and structured JSON output, available via API.
Specifications
- Input
- text, documents
- Output
- text
- Context size
- 128 000 tokens
- Max answer length
- 102 400 tokens
- Capabilities
- function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Mistral Large 2?
Mistral Large 2 is Mistral AI's flagship model, version mistral-large-2407. It is a proprietary model with available weights, strong in reasoning, coding, structured JSON generation, and dialogue. In BotHub, it is accessible without a VPN or foreign card: pay with a Russian card in rubles only for the tokens you use, which do not expire. With over 250 other neural networks in one window, you can compare Mistral Large 2's responses with other models and switch without rewriting your integration: a single OpenAI-compatible API provides them via one key. Correspondence is encrypted via AES-GCM and not saved. Companies have access to contracts, invoices, electronic document management, and an admin panel with limits. Developers can use the model for writing and refactoring code, debugging, and code reviews; analysts for extracting data into strict JSON; product teams for chatbots and support assistants; and those working with logic and calculations for step-by-step reasoning and hypothesis testing.Frequently asked questions about Mistral Large 2
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.
mistral-large is a text model from Mistral AI. It is suitable for text generation and editing, working with code (writing, refactoring, debugging), analytics, and document parsing: files are transmitted natively without prior recognition. It supports function calling, making it easy to integrate into services and workflows.
Up to 102,400 tokens per response — enough for a large document, detailed instruction, or an entire large code module. The total context window is 128,000 tokens, which includes your request with attached files and the model's response.
A separate reasoning mode is not noted in our data. However, you can ask the model to break down a task step-by-step directly in the prompt: describe logic, iterate through options, and verify calculations. For complex reasoning, it is easy to switch to another model in the same window in BotHub.
Yes, both. Function calling allows the model to access your tools: databases, search, calculators. Structured output returns results strictly in JSON according to a specified schema, which is convenient for agents and parsing. Everything works via the unified OpenAI-compatible BotHub API.
mistral-large does not accept images, audio, or video as input. However, it has native document support: the file is sent to the model as is, and it parses the structure, tables, and wording. For images, audio, and video in BotHub, use separate models in the same interface.
The model's language performance is not specifically described in our data, so we cannot promise specific quality for Russian — please test it with your task. It's fast: in BotHub, you can send the same request to multiple models in one window and compare the responses.