MiniMax-M2.5 is a general-purpose large language model developed by MiniMax, designed to power a wide spectrum of natural language applications from intelligent chatbots and virtual assistants to automated content generation and document analysis pipelines.
The flagship general-purpose language model from MiniMax. Delivers superior instruction-following, nuanced reasoning, and high-fidelity content generation. Designed for workloads where response quality and contextual depth are the primary objectives.
A throughput-optimized variant engineered for latency-sensitive applications. Achieves significantly faster time-to-first-token and higher requests-per-second capacity, making it the go-to choice for live user interactions and high-traffic services.
Power multi-turn, context-aware conversations for customer service, support automation, and virtual assistant platforms with natural, coherent dialogue management.
Automate the creation of articles, marketing copy, product descriptions, social media posts, and long-form editorial content at scale without sacrificing quality.
Summarize, classify, extract key information from, and answer questions about contracts, reports, research papers, and enterprise documents using extended context.
Serve as the reasoning backbone for autonomous agents, enabling complex task decomposition, tool selection, multi-step planning, and iterative self-correction cycles.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
MiniMax M2.5 is documented as suited to coding and software engineering assistance, including SWE-bench-style tasks.
It is also documented for search and tool-calling workflows, including agentic tool use and information-retrieval tasks.
A third documented use case is office productivity work, such as document- and spreadsheet-oriented workflows.
It is also described as appropriate for general conversational assistant use.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "minimax/m2-5-20260218", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "minimax/m2-5-20260218", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"minimax/m2-5-20260218","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 22.8 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| SWE-bench Verified | 80.2% | Resolving verified real GitHub issues | Source | September 21, 2026 |
| AIME 2025 | 86.3 | Competition mathematics (AIME), 2025 | Source | September 21, 2026 |
| GPQA Diamond | 85.2 | Google-proof graduate science questions (hardest subset) | Source | September 21, 2026 |
| Humanity's Last Exam | 19.4 | Expert-level questions across many domains | Source | September 21, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
MiniMax-M2.5 This page | Coding + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
MiniMax-M2.5 has a 204,800 tokens context window and can return up to 204,800 tokens.
MiniMax-M2.5 takes image, text as input and returns text.
Use minimax/m2-5-20260218 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
MiniMax-M2.5 is priced at input $0.39 / 1M tokens, output $1.56 / 1M tokens, cached input $0.39 / 1M tokens.
Yes, MiniMax-M2.5 can stream responses as they are generated.
Yes, MiniMax-M2.5 accepts image input alongside text.
MiniMax-M2.5 was built by MiniMax.
Yes, it supports function calling, tool use, and parallel tool calls, along with structured output formatting.
