MiniMax M2.7
minimax/minimax-m2.7MiniMax M2.7 is a text model with reasoning, tool support, open weights, and token-based pricing.
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 18 Mar 2026 · checked 3 Oct 2026.
About it
MiniMax M2.7 is a text-only model from MiniMax, released March 18, 2026. It supports reasoning and tools, accepts and produces text, and has open weights. Its 204,800-token context can hold long inputs, while output can reach 176,947 tokens. Pricing is $0.21 per million input tokens and $0.84 per million output tokens.
Review
In a line
MiniMax M2.7 is a text model made by MiniMax. It accepts text and returns text, with support for reasoning and tools.
What it's good at
This model suits tasks that need step-by-step reasoning or tool use. Its open weights make it relevant to projects choosing open-weight models.
It works with text only, so it is aimed at language tasks rather than image input or image output.
Price in plain terms
Input costs $0.21 per million tokens. Output costs $0.84 per million tokens.
Reading a page of about 500 words costs about 0.014 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.063 cents. At that same size, 1,000 turns cost about $0.63.
Context and limits
The model can take up to 204,800 tokens in one context, so a single request can include a large amount of text. It can produce up to 176,947 tokens in one answer.
Its stated inputs, outputs, and modality are text.
Ways to use it
Use MiniMax M2.7 in software that supports its model ID, minimax/minimax-m2.7. Its Hugging Face ID is MiniMaxAI/MiniMax-M2.7.
The model includes tool support and has open weights, so it can fit projects selecting MiniMax models or models with tool calling.
Verdict
Choose M2.7 when your work is text-based and benefits from reasoning, tool calls, or a very large context. Its open weights may also suit teams that need that model format.
Look elsewhere if you need image understanding or image creation. The token charges are low for the example workloads listed, but your bill rises with both the text you send and the text it generates.
Verdict
MiniMax M2.7 is a sensible pick for language workflows that need long inputs, reasoning, and tool support. The open-weight release gives teams another way to work with the model, while the published model and Hugging Face IDs make it identifiable in compatible software. Its pricing is easy to estimate for small chats, though output costs more than input. It is not a fit for image-based tasks, since the listed modality covers text only.
- Reasoning and tool support
- 204,800-token context window
- Open weights available
- Low example cost for 1,000 chat turns
- Text only; no listed image support
- Output tokens cost more than input tokens
- Maximum answer length is capped at 176,947 tokens
- Pick it ifText workflows needing reasoning or tool calls; Projects using open-weight models; Large text prompts
- Skip it ifYou need image input or image output; Your software cannot use the listed model IDs
Pricing
Input is $0.21 per million tokens and output is $0.84 per million tokens. The examples provided estimate about 0.014 cents for one page, 0.063 cents for one chat turn, and $0.63 for 1,000 such turns.
Questions
What is MiniMax M2.7?
It is a text model from MiniMax with reasoning, tool support, and open weights.
How much does MiniMax M2.7 cost?
Input costs $0.21 per million tokens. Output costs $0.84 per million tokens.
How much text can it read at once?
Its context window is 204,800 tokens.
How long can one answer be?
The maximum output is 176,947 tokens.
Can MiniMax M2.7 handle images?
The listed inputs, outputs, and modality are text, with no image support specified.
Where can I identify the model?
Its model ID is minimax/minimax-m2.7, and its Hugging Face ID is MiniMaxAI/MiniMax-M2.7.
Does it support tools?
Yes. Tool support is listed for MiniMax M2.7.