Does Magistral Medium 1.2 support zero data retention?

Not self-serve on Pay-as-you-go. Zero data retention for Magistral Medium 1.2 is available on request via Opper Enterprise — contact hello@opper.ai.

Where is Magistral Medium 1.2 hosted?

Magistral Medium 1.2 runs on EU hosting routes through the Opper gateway, with contractually enforced residency on supported providers.

Is Magistral Medium 1.2 trained on prompts and outputs?

No. Every hosting route for Magistral Medium 1.2 is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for Magistral Medium 1.2?

Yes. A GDPR Data Processing Agreement is available on at least one route serving Magistral Medium 1.2. Contact hello@opper.ai to receive it.

Magistral Medium 1.2

by Mistral

Magistral Medium 1.2 is Mistral's September 2025 update to its frontier-class reasoning model, combining advanced step-by-step reasoning with vision. The 1.2 release adds image understanding and reports notable gains on math and coding benchmarks over the original June 2025 Magistral. The model is built for transparent, interpretable reasoning across multiple steps and professional domains such as legal research, financial analysis, and software development. Its multi-step chains of thought are designed to be traceable, which suits high-stakes decision-making and complex problem solving. With a 128k token context window and multilingual reasoning across languages including English, French, Spanish, German, Italian, Arabic, Russian, and Simplified Chinese, Magistral Medium 1.2 pairs visual analysis with structured logical thinking.

Open in playground Try in chat Compare

Key info

Input

Output

Features

Context window

128K

Max output

Input price

$2.00 /1M

Output price

$5.00 /1M

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

Available routes

Magistral Medium 1.2 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
Mistral	EU	Enterprise	No	$2.00	$5.00

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

Data handling per route

Each route hosting Magistral Medium 1.2 has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

Mistral — France🇫🇷

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

Zero data retention: Available via Opper Enterprise contract.
Training: No training on customer data.
Logging: Abuse monitoring (30-day retention)
Third-party access: Provider may share with subprocessors / partners
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#190 of 531 LLMs

TierEfficient

Output speed40 tok/s

First token0.61s

Intelligence Index20.1

Coding Index21.7

Math Index82.0

Reasoning & knowledge

MMLU-Pro

82%

GPQA Diamond

74%

Humanity's Last Exam

10%

Long-context reasoning

51%

Coding

LiveCodeBench

75%

SciCode

39%

Agentic & tool use

Terminal-Bench Hard

13%

τ²-Bench Telecom

52%

Math & instruction following

AIME 2025

82%

IFBench

43%

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

Get started

Call Magistral Medium 1.2 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: "mistral/magistral-medium-2509",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);