Does MiniMax M3 support zero data retention?

Not self-serve on Pay-as-you-go. Zero data retention for MiniMax M3 is available on request via Opper Enterprise — contact hello@opper.ai.

Where is MiniMax M3 hosted?

MiniMax M3 is available on multi-region hosting through the Opper gateway. See the per-route table for details.

Is MiniMax M3 trained on prompts and outputs?

No. Every hosting route for MiniMax M3 is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for MiniMax M3?

Yes. A GDPR Data Processing Agreement is available on at least one route serving MiniMax M3. Contact hello@opper.ai to receive it.

MiniMax M3

by MiniMax

MiniMax M3 is an open-weight multimodal model released June 2026, combining frontier coding performance with a 1M-token context window and native image and video understanding. It is positioned as the first open-weight model to bring frontier coding, million-token context, and native multimodality together in one architecture. The model is built around roughly 428 billion parameters with about 23 billion activated per token, using MiniMax Sparse Attention to cut per-token compute at 1M-token scale to roughly one-twentieth that of the previous generation, with more than 9 times faster prefill and more than 15 times faster decoding versus M2. It undergoes mixed-modality training from the start for deeper fusion across text, image, and video. M3 targets long-horizon agentic and coding tasks, scoring 59% on SWE-Bench Pro and 83.5 on BrowseComp. It supports multiple reasoning modes through the thinking parameter and is designed for tool use, computer use, and document and chart parsing across extended contexts.

Open in playground Try in chat Compare

Key info

Input

Output

Features

Context window

Max output

131K

Input price

$0.60 /1M

Output price

$2.40 /1M

Zero data retention via Enterprise
No training by default
GDPR DPA available

Available routes

MiniMax M3 runs on 1 route through the Opper gateway. Compare residency, ZDR, and training posture at a glance — full data-handling detail per route below.

Provider	Region	Zero data retention	Training	Input	Output
MiniMax	Multi	Enterprise	No	$0.60	$2.40

Training posture across routes: No training on prompts by default.

Data handling per route

Each route hosting MiniMax M3 has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

MiniMax — Singapore🇸🇬

Zero data retention is available via Opper Enterprise contract. No training on customer data. US; SCCs; DPA available.

Zero data retention: Available via Opper Enterprise contract.
Training: No training on customer data.
Logging: Abuse monitoring
Third-party access: None disclosed
GDPR DPA: DPA available
Transfer mechanism: SCCs

Get started

Call MiniMax M3 through the Opper gateway with one API key. Let your coding agent set it up, or call it directly — Opper is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.

Set it up with your agent

Copy this and paste it into your coding agent — Claude Code, Cursor, Codex, and more — 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/m3",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);