MiniMax M2

Provider: MiniMax CN

No longer listed on OpenRouter since 2026-11-05. The figures below are the last recorded values.
Input price
$0.26
/ 1M tokens
Output price
$1.02
/ 1M tokens
Cache price
$0.03
/ 1M tokens (read)
Context
205K tokens
Max output 131K tokens

Overview

MiniMax M2 is a general-purpose model from MiniMax (CN), positioning itself as a capable mid-tier offering. It targets workloads requiring structured output, complex reasoning, and extensive memory. Its large context window and specific capabilities make it suitable for various advanced applications.

Strengths

  • Very large context window of 204800 tokens supports extensive input processing.
  • Exceptional maximum output length of 196608 tokens for comprehensive responses.
  • Robust function-calling and JSON-mode capabilities enable reliable structured interactions.
  • Includes "thinking" capability, beneficial for more complex logical tasks.
  • Extremely low cache read price ($0.03/1M tokens) for cost-efficient state management.

Best use cases

  • Developing AI agents that require precise function invocation and structured data output.
  • Summarizing or generating long-form content from vast amounts of text data.
  • Building sophisticated conversational applications leveraging cost-effective cache reads.
  • Handling complex logical reasoning tasks over large context windows.

Pricing position

MiniMax M2's input price of $0.255/1M tokens positions it above MiniMax M2.5 ($0.15/M) and MiniMax-01 ($0.2/M) but below other M2 series models. Its $1/1M output price is competitive, significantly undercutting GPT-5 Mini ($2/M), Claude 3 Haiku ($1.25/M), and Gemini 3.1 Flash Lite ($1.5/M), although DeepSeek V3.2 offers a much lower output cost at $0.378/M.

Caveats

The model does not include multimodal capabilities like vision support, which might restrict its use in certain applications. While its context window is substantial, MiniMax-01 offers a dramatically larger context of 1M tokens at a lower input price. Additionally, there is no free tier provided, which could be a barrier for initial experimentation.

Notes drafted by an LLM from the figures on this page (2026-05-21) and rewritten when the price changes.

Key facts

Max output
131K tokens
Tracked since
2026-05-09
Last updated
2026-11-05

Features

Function callingReasoning

Price history

  • 2026-11-05 No longer listed

Monthly cost estimate

Scale Input (M) Output (M) Monthly
0.1M (hobby) 0.1 0.05 $0.08
1M (small app) 1 0.5 $0.77
10M (mid scale) 10 5 $7.65
100M (large) 100 50 $76.50

※ Cache and batch discounts not factored in. Detailed calculator

Official resources

Other models from MiniMax

Similar-priced models

Source: OpenRouter API