MiniMax M2
Neural Network
MiniMax M2 is a compact model from MiniMax for end-to-end development and agent scenarios, available via MiniMax M2 API.
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works MiniMax M2?
MiniMax M2 is a compact language model from MiniMax, designed for end-to-end development and agent scenarios. Out of 230 billion parameters, 10 billion are activated at each step, allowing the model to maintain near-state-of-the-art reasoning while remaining computationally efficient. Its strengths include coding from task definition to final result, general reasoning, and long chains of thought. In BotHub, the model works without a VPN or foreign card: you pay with a Russian card for consumed tokens, and the balance does not expire. Over 250 models are available in one window, allowing you to compare answers in a few clicks, and a unified OpenAI-compatible API lets you replace the model in a multi-model system without rewriting the integration. Correspondence is encrypted, and companies have access to contracts, invoices, and an admin panel with limits. Developers use MiniMax M2 for code generation and refactoring, debugging, and analyzing repositories; analysts use it for step-by-step reasoning over data; and product teams build agents for multi-step tasks.Frequently asked questions about MiniMax M2
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.
minimax-m2 is a text model from MiniMax: it writes and edits code, analyzes long documents, solves reasoning tasks, and accepts images as input. A context of up to 196,608 tokens allows you to keep a large repository, technical documentation, or entire correspondence in one dialogue.
The single response limit is 176,947 tokens, which is very high for text models: enough for a bulky code module, a detailed technical analysis, or an entire long document. The total dialogue volume is limited by the 196,608-token context.
Yes, the reasoning mode in minimax-m2 is confirmed: the model first builds a solution path and then provides an answer. This significantly helps with math, logic, code debugging, and multi-step tasks where it is important to arrive at the result sequentially rather than guessing.
Yes. minimax-m2 supports function calling and structured JSON output, making it convenient for building agents and pipelines. In BotHub, the model is available via a unified OpenAI-compatible API: you describe the tool schema once and switch models without rewriting the integration.
Images and documents — yes: you can send an interface screenshot, diagram, chart, or PDF and ask for an analysis. We have no data on audio or video input, so for voiceovers or video processing, choose a specialized model in BotHub.
We have no confirmed data on the quality of work with specific languages, so it is better to check in practice: ask a couple of typical questions and attach your document. If the answer is not satisfactory, other models are available in the same BotHub window — you can compare them in a minute.