FlagshipMiniMax

MiniMax M2.5

MiniMax M2.5 is MiniMax's flagship text generation model with a 128K token context window and tool calling capabilities.

Context 128K
Tier Flagship
Tools Supported
Input from
$0.118 / 1M tokens
across 3 providers

API Pricing

Cheapest on OpenRouter 51% below avg
ProviderInput / 1MOutput / 1MSpeedTTFTUpdated
$0.118$0.99070.7 t/s2.0s4/14/2026
$0.300$1.2070.7 t/s2.0s4/14/2026
$0.300$1.2070.7 t/s2.0s4/14/2026

Prices updated daily. Last check: 4/14/2026

Model Details

General

Creator
MiniMax
Family
MiniMax
Tier
Flagship
Context Window
128K
Modalities
Text

Capabilities

Tool Calling
Yes
Open Source
No
Subtypes
Chat Completion
Aliases
minimax-m2-5, minimax-m2-family

Strengths & Limitations

  • 128,000 token context window for processing lengthy documents
  • Tool calling support enables API integrations and structured outputs
  • 58.52 tokens per second generation speed for responsive interactions
  • Flagship tier positioning within MiniMax's model family
  • Specialized capabilities for Chinese language processing
  • Chat completion interface for conversational applications
  • Developed by MiniMax with focus on regional market needs
  • Text-only modality without image or multimodal input support
  • Proprietary model with no open source weights available
  • Limited global availability compared to major cloud providers
  • Smaller context window than some competing flagship models
  • Time to first token of 1,386ms may impact real-time applications

Key Features

128,000 token context window
Tool calling with external API integration
Chat completion interface
Text generation at 58.52 tokens per second
Structured output capabilities
Chinese language optimization
Conversational AI support
Multi-turn dialogue handling

About MiniMax M2.5

MiniMax M2.5 is the flagship model from Chinese AI company MiniMax, representing their primary offering for general-purpose text generation and reasoning tasks. As a proprietary model positioned at the top of MiniMax's model lineup, M2.5 targets complex conversational AI and business applications requiring sophisticated language understanding. The model operates with a 128,000 token context window and supports text-only interactions through chat completion interfaces. M2.5 includes tool calling functionality, enabling integration with external APIs and structured workflows. Performance benchmarks show the model generates approximately 58.52 tokens per second with a time to first token of 1,386 milliseconds, indicating balanced throughput characteristics for interactive applications. M2.5 serves organizations requiring advanced Chinese and multilingual language processing capabilities, particularly in markets where MiniMax has established partnerships. The model competes in the flagship tier alongside other major language models, though with a more focused regional emphasis compared to globally-distributed alternatives.

Common Use Cases

MiniMax M2.5 suits organizations requiring flagship-tier language processing with particular emphasis on Chinese markets and applications. The 128K context window makes it appropriate for document analysis, lengthy conversational sessions, and complex reasoning tasks that require maintaining context across extended interactions. Tool calling capabilities enable integration into business workflows, customer service automation, and agentic applications that need to interact with external systems. The model's regional optimization makes it valuable for companies operating in Chinese-speaking markets or requiring specialized understanding of regional language nuances and cultural context.

Frequently Asked Questions

How much does MiniMax M2.5 cost per million tokens?

MiniMax M2.5 pricing varies by provider and region. Check the pricing table above for current rates across all available providers offering this model.

What is MiniMax M2.5 best used for?

MiniMax M2.5 is best suited for complex conversational AI applications, document processing requiring its 128K context window, and business workflows that leverage its tool calling capabilities. It's particularly valuable for organizations operating in Chinese-speaking markets or requiring regional language expertise.

Does MiniMax M2.5 support multimodal inputs like images?

No, MiniMax M2.5 is a text-only model that supports chat completion interfaces but does not process images or other multimedia inputs. It focuses exclusively on text generation and understanding tasks.