MiniMax M2.5

minimax/minimax-m2.5
Reasoning
by MiniMax · 2026-02-12

MiniMax M2.5 — SOTA productivity LLM with strong coding + agentic capabilities, 200k context, ~60 tps output.

ctx204.8K tokens
Max output2K
Inputtext
Outputtext
p50 TTFT2.00 s
INPUT$0.30/ 1M tokens
OUTPUT$1.20/ 1M tokens
p50 TTFT2.00 s7d
p95 TTFT2.26 s7d
TRAFFIC14.7Ktokens / 7d

MiniMax M2.5 is a large language model developed by Minimax and made available through OrcaRouter's API. It is designed to process natural language inputs and generate text outputs. The model is…

What is MiniMax M2.5?

Who should use MiniMax M2.5?

Key specifications of MiniMax M2.5

How to access MiniMax M2.5 via OrcaRouter

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key="$ORCAROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="minimax/minimax-m2.5",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • max_completion_tokens
  • reasoning
  • reasoning_split
  • stream
  • temperature
  • top_p

Pricing

Input / 1M tokens$0.300
Output / 1M tokens$1.20
Cache read / 1M$0.030
Cache write / 1M$0.375
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $5.70 · With prompt caching $4.76

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.000606

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
2.00 s
Output speed
95.0 tok/s
p95 TTFT
2.26 s
Error rate
0%

Public benchmarks

37.4
AA Coding
Better than 45% of models compared
#57 of 106
41.9
AA Intelligence
Better than 45% of models compared
#60 of 110
GPQA Diamond
84.8
Humanity's Last Exam
19.1
IFBench
71.6
Long-Context Recall
66.0
SciCode
42.6
TerminalBench Hard
34.8
τ²-Bench
95.3
Source: artificialanalysis.ai

How it compares

MiniMax M2.5MiniMax M3MiniMax M2.7MiniMax M2.7 highspeed
Input $/M$0.30$0.30$0.30$0.60
Output $/M$1.20$1.20$1.20$2.40
Context205K1.0M205K205K
Quality7/109/108/108/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

What is the cost to use MiniMax M2.5 per token?
Input tokens are billed at $0.30 per 1 million tokens, and output tokens at $1.20 per 1 million tokens. OrcaRouter adds no markup; you pay the provider rate.
What is the context window size of MiniMax M2.5?
It supports a context window of 204800 tokens, allowing you to include very long prompts.
What are the main strengths of MiniMax M2.5?
Its large context window (204800 tokens) and strong performance on task-oriented benchmarks (τ²-Bench score 95.3) are key strengths. It is also competitively priced for long-context tasks.
How does MiniMax M2.5 compare to other long-context models?
It offers a larger context than many alternatives at a lower per-token cost. However, it is text-only and has a max output of 2048 tokens. Direct benchmark comparisons beyond τ²-Bench are not provided.
How can I call MiniMax M2.5 via an OpenAI-compatible API?
Use OrcaRouter's base URL https://api.orcarouter.ai/v1 with model ID "minimax/minimax-m2.5". Provide your OrcaRouter API key in the Authorization header. Standard chat completions parameters are supported.
Does MiniMax M2.5 handle multimodal inputs?
No, it is a text-only model. It does not accept images, audio, or video.
What is the maximum output length of MiniMax M2.5?
The maximum output is 2048 tokens per request.
Is there any markup on pricing when using OrcaRouter?
No, OrcaRouter bills at the provider rate with zero markup for MiniMax M2.5.
What is τ²-Bench and what does a score of 95.3 indicate?
τ²-Bench evaluates task-oriented performance. A score of 95.3 indicates that MiniMax M2.5 performs well on such tasks relative to other models, but it is just one benchmark.
Can I use MiniMax M2.5 for code generation?
Yes, as a text model it can generate code, but its maximum output is 2048 tokens, which may be limiting for large code blocks. Performance on coding benchmarks is not provided.

Embed this badge

MiniMax: MiniMax M2.5$0.30/M in2000ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/minimax/minimax-m2.5" target="_blank"> <img src="https://www.orcarouter.ai/embed/minimax/minimax-m2.5.svg" alt="MiniMax: MiniMax M2.5 on OrcaRouter" /> </a>
Markdown [![MiniMax: MiniMax M2.5](https://www.orcarouter.ai/embed/minimax/minimax-m2.5.svg)](https://www.orcarouter.ai/models/minimax/minimax-m2.5)

Model card as data

GET /api/public/models/minimax/minimax-m2.5Open
Machine-readable:/llms.txt/llms-full.txt