Models·MiniMax

MiniMax M2: Pricing, Context and Tool Support

minimax/minimax-m2

MiniMax M2 is a text model for coding, reasoning, and tool-based agent workflows.

$0.30Input, per million tokens
$1.20Output, per million tokens
205KContext window
177KLongest answer

205K tokens of context — about 153,600 words, on a scale that ends at 2M.

What it can do

  • Reasoning
  • Tool use
  • Open weights
  • Takes text
  • Gives text

Released 23 Oct 2025 · checked 3 Oct 2026.

About it

MiniMax M2 is a text-only model from MiniMax, released October 23, 2025. It supports reasoning and tools, with open weights available under the MiniMaxAI/MiniMax-M2 identifier. It accepts text, writes text, and has a 204,800-token context window. Input costs $0.3 per million tokens and output costs $1.2 per million tokens.

Review

In a line

MiniMax M2 is a large language model from MiniMax for coding, general reasoning, and workflows where software tools are part of the task. It takes text in and returns text.

What it's good at

The model is aimed at end-to-end coding and agent workflows, meaning tasks that can involve several steps and tool calls. It supports reasoning and tools, so a compatible application can let it work with those tools.

MiniMax M2 has 10 billion activated parameters out of 230 billion total. Its weights are open, and the model is listed as MiniMaxAI/MiniMax-M2.

Price in plain terms

Input costs $0.3 per million tokens, and output costs $1.2 per million tokens. Tokens are small pieces of text counted for billing.

At the stated rates, reading a 500-word page costs about 0.019 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.09 cents. One thousand such turns cost about $0.90.

Context and limits

The context window is 204,800 tokens, so one request can include a large amount of text for the model to consider. A single answer can contain up to 176,947 tokens.

MiniMax M2 handles text input and produces text output.

Ways to use it

Use MiniMax M2 in an app or chat tool that offers the model, with tool support available where that software connects tools to it. Its model ID is minimax/minimax-m2.

Because its weights are open, it can also be considered alongside other open-weight models. MiniMax M2 belongs to the minimax family.

Verdict

Choose MiniMax M2 if your work centers on text coding, multi-step reasoning, or tool-assisted agents and you want open weights. Its large context can hold long instructions or documents, while the listed token rates keep small and repeated chats inexpensive.

Look elsewhere if you need image understanding or image creation, since this model is text-only. Also check that the app or service you plan to use supports its model ID and tool behavior before relying on it.

Verdict

MiniMax M2 fits people building or selecting text-based coding and agent workflows, especially when open weights matter. Its low input rate and modest output rate make frequent conversations affordable at the stated usage pattern. The wide context window is useful for long prompts, but the model does not cover image tasks. Before choosing it, confirm that your chosen software exposes MiniMax M2 and supports the tools you need; the facts do not name specific apps, integrations, or hosted services.

  • Open weights are available under MiniMaxAI/MiniMax-M2.
  • Supports reasoning and tool use for coding and agent workflows.
  • 204,800-token context window.
  • About $0.90 for 1,000 stated chat turns.
  • Text input and output only.
  • Single responses are limited to 176,947 tokens.
  • No specific apps, integrations, or hosted services are listed.
  • Pick it ifText coding and multi-step agent workflows; Users comparing open-weight models; High-volume chat at low token cost
  • Skip it ifYou need image input or image output.

Pricing

Input is $0.3 per million tokens and output is $1.2 per million tokens. The stated example costs about 0.019 cents to read one page, about 0.09 cents per chat turn, and about $0.90 for 1,000 such turns.

Questions

What is MiniMax M2?

MiniMax M2 is a text-only large language model from MiniMax for coding, reasoning, and tool-based agent workflows.

Can MiniMax M2 use tools?

Yes. Tools are supported, although the facts do not name particular tools or integrations.

Is MiniMax M2 open weight?

Yes. Its listed Hugging Face identifier is MiniMaxAI/MiniMax-M2.

How much text can it consider?

Its context window is 204,800 tokens.

How long can one answer be?

The maximum output is 176,947 tokens.

What does MiniMax M2 cost?

Input costs $0.3 per million tokens and output costs $1.2 per million tokens.

Does it handle images?

No image capability is listed. Its inputs and outputs are text.

When was MiniMax M2 released?

The model was released on October 23, 2025.

On these boards

More from MiniMax