MiniMax logo

MiniMax M2.5: Pricing, Context Window & Benchmarks

by MiniMax

MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.

Chat with MiniMax M2.5
Input Price
$0.30/1M tokens
Output Price
$1.20/1M tokens
Intelligence
42.0
Coding
37.4

What you can do with MiniMax M2.5

Everyday Q&A and clear explanations

Writing help (emails, posts, summaries)

Idea generation and brainstorming

Learning support with step-by-step guidance

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Benchmark Highlights

2 tests
GPQA
84.8%
MMLU Pro
N/A
LiveCodeBench
N/A
Math 500
N/A
AIME 2025
N/A
HLE
19.1%
Metric Value
Provider MiniMax
Context Window 196,608 tokens
Input Price $0.30/1M tokens
Output Price $1.20/1M tokens
Release Date Feb 12, 2026
Modalities text
Capabilities N/A

Compare MiniMax M2.5 to other models

See how it stacks up on price, quality, and overall performance.

Frequently asked questions

What is MiniMax M2.5 good for?

Use MiniMax M2.5 for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does MiniMax M2.5 cost?

Pricing is based on usage. Current rates are $0.30/1M tokens for input and $1.20/1M tokens for output.

Can I try MiniMax M2.5 for free?

Yes. You can start a chat instantly and test the model before deciding on a plan.

Does MiniMax M2.5 support images or audio?

MiniMax M2.5 focuses on text-based tasks.

Benchmarks and pricing are sourced from Artificial Analysis where available. OpenRouter specs are used as a fallback.