MiniMax M2.7

MiniMax · released 18 Mar 2026

Input$0.21per 1M tokens
Output$0.84per 1M tokens
Context205Ktokens
WeightsOpen
context on a 1K–2M scale

About

MiniMax M2.7 is a text model from MiniMax, described as a large language model for autonomous productivity and continuous improvement. Its summary says it is designed to participate in its own evolution and refers to agentic capabilities through multi-agent work, without specifying further details. It accepts text and returns text. Tool use and reasoning are listed, as are open weights. Its context window is 204,800 tokens, and its maximum output is 176,947 tokens. Input costs 0.21 USD per 1M tokens; output costs 0.84 USD per 1M tokens. The model was released on 2026-03-18. Its Hugging Face ID is MiniMaxAI/MiniMax-M2.7.

Who it is for

It suits teams exploring text-based productivity or multi-agent workflows that use tool calls and reasoning. Open weights may be relevant for teams considering that access model.

What is good

  • Open weights listed
  • Tool use and reasoning listed
  • 176,947-token maximum output
  • 204,800-token context window

What to know first

  • Text input and output only
  • Output costs 0.84 USD per 1M tokens

Verdict

MiniMax M2.7 brings tool use, reasoning and open weights to a text-only model positioned for productivity and multi-agent work. Its listed context, output limit and token rates help define fit for a workload.

Details

Lab
MiniMaxopenrouter.ai · 3 Oct 2026
Context
205Kopenrouter.ai · 3 Oct 2026
Input price
$0.21 / 1Mopenrouter.ai · 3 Oct 2026
Output price
$0.84 / 1Mopenrouter.ai · 3 Oct 2026
Max output
176,947 tokensopenrouter.ai · 3 Oct 2026
Inputs
textopenrouter.ai · 3 Oct 2026
Open weights
Yesopenrouter.ai · 3 Oct 2026

More newest models

See the list