When was MiniMax M2.5 released?

MiniMax released MiniMax M2.5 on 12 Feb 2026. It is available through the Opper gateway.

Does MiniMax M2.5 support zero data retention?

Yes. MiniMax M2.5 is available with zero data retention on GreenPT, Novita, TensorX through the Opper gateway. Other routes carry zero data retention with abuse monitoring; see each route's window in the data-handling section.

Where is MiniMax M2.5 hosted?

MiniMax M2.5 runs on both EU and US hosting routes through the Opper gateway. Pick the route that matches your residency requirement.

Is MiniMax M2.5 trained on prompts and outputs?

No. Every hosting route for MiniMax M2.5 is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for MiniMax M2.5?

Yes. A GDPR Data Processing Agreement is available on at least one route serving MiniMax M2.5. Contact hello@opper.ai to receive it.

MiniMax M2.5

by MiniMax

Released 51 days after MiniMax M2.1 · All MiniMax releases

MiniMax M2.5 is a 229-billion-parameter Mixture-of-Experts model released February 2026, trained extensively with reinforcement learning across hundreds of thousands of real-world environments and more than 10 programming languages including Go, C, C++, TypeScript, Rust, Kotlin, Python, Java, JavaScript, PHP, Lua, Dart, and Ruby. It combines strong reasoning with agentic and instruction-following capability. M2.5 scores 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, and leads on multi-turn function calling with 76.9% on BFCL. It completes SWE-Bench evaluations about 37% faster than M2.1 while matching Claude Opus 4.6's throughput, and uses roughly 20% fewer interaction rounds than its predecessor for more efficient task decomposition. The model supports extended chain-of-thought reasoning for complex problems and is strong at professional office work, reaching a 59.0% average win rate against competing models on workplace tasks. Recommended sampling settings are temperature 1.0, top_p 0.95, and top_k 40.

Call MiniMax M2.5 on Opper with the OpenAI SDK. Sign up without a credit card. Get started

Get your API key Try in chat Compare

Key info

Input

Output

Features

Context window

66K

Max output

66K

Input price

$0.30 /1M$0.06 cache read

Output price

$1.20 /1M

Released

12 Feb 2026

EU residency available
US residency available
Zero data retention on pay-as-you-go
No training by default
GDPR DPA available

Available routes

MiniMax M2.5 runs on 4 different routes through the Opper gateway. Compare residency, zero data retention and training posture at a glance, with full data-handling detail per route below. In the EU it is served by GreenPT and TensorX (EU multi-region). See every model hosted in the EU.

Provider	Region	Training	Input	Output	Cache read
DigitalOcean	US	No	$0.30	$1.20	$0.06
GreenPT	EU	No	$0.38	$1.51	Input price
Novita	US	No	$0.30	$1.20	Input price
TensorX	EU	No	$0.30	$1.20	$0.08

Uptime and availability

One of MiniMax M2.5's 4 routes is on the monitoring board, so the figure below is that provider's availability rather than the model's.

100%route uptime, last 30 days

Its other routes aren't measured here yet.

Name a second model in the same request and the gateway tries it on retriable errors, so a busy hour never has to reach your users. Set up a fallback chain.

Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. 3 further routes are not on the board yet. See uptime for every provider Opper monitors.

Data handling per route

Each route hosting MiniMax M2.5 has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

DigitalOcean — United States🇺🇸

Zero data retention with abuse monitoring: nothing is logged or used for training; content flagged by abuse monitoring is held for review. No training on customer data. US; dpf sccs; DPA available.

Zero data retention: With abuse monitoring, flagged content held.
Training: No training on customer data.
Logging: None
Abuse monitoring: Always on, holds flagged content, stored in United States
Caching: Content cached for replay
Subprocessor access: No subprocessor reads content
GDPR DPA: DPA available
Transfer mechanism: dpf sccs

GreenPT — EU🇪🇺

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EU; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: Keeps nothing
Caching: Content cached for replay
Subprocessor access: Subprocessors may read content
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Novita — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: On by default, keeps nothing
GDPR DPA: DPA available
Transfer mechanism: SCCs

TensorX — European Union🇪🇺

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EU; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
Caching: Content cached for replay
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Benchmarks

Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.

Global rank#173 of 665 LLMs

TierEfficient

Intelligence Index22.8

Reasoning & knowledge

GPQA Diamond

85%

Humanity's Last Exam

21%

Long-context reasoning

73%

Agentic & tool use

Terminal-Bench Hard

35%

τ²-Bench Telecom

95%

Math & instruction following

IFBench

72%

See full leaderboard →Benchmarks via Artificial Analysis · View on AA

Get started

Call MiniMax M2.5 through the Opper gateway with one API key. Opper is drop-in compatible with the OpenAI, Anthropic and Google AI SDKs.

MiniMax M2.5 is a premium model on Opper. Sign up needs no credit card: you get an API key straight away and the free models work in the playground and the API. Add a card to use premium models, pay-as-you-go with no minimum.

Set it up with your agent

Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up Opper for you.

Or call it directly

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.OPPER_API_KEY,
5  baseURL: "https://api.opper.ai/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "minimax-m2.5",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);