Does Qwen 3 Next 80B A3B Thinking support zero data retention?

Not self-serve on Pay-as-you-go. Zero data retention for Qwen 3 Next 80B A3B Thinking is available on request via Opper Enterprise — contact hello@opper.ai.

Where is Qwen 3 Next 80B A3B Thinking hosted?

Qwen 3 Next 80B A3B Thinking runs on both EU and US hosting routes through the Opper gateway. Pick the route that matches your residency requirement.

Is Qwen 3 Next 80B A3B Thinking trained on prompts and outputs?

No. Every hosting route for Qwen 3 Next 80B A3B Thinking is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for Qwen 3 Next 80B A3B Thinking?

Yes. A GDPR Data Processing Agreement is available on at least one route serving Qwen 3 Next 80B A3B Thinking. Contact hello@opper.ai to receive it.

Qwen 3 Next 80B A3B Thinking

by Alibaba

Qwen 3 Next 80B A3B Thinking is an open-weight, sparse mixture-of-experts model that pairs the efficiency of roughly 3B active parameters with extended reasoning. Unlike the Instruct variant, it generates reasoning traces before its final answer, exposing its problem-solving steps across math, logic, coding, and planning tasks. With a 128K context window and a sparse MoE design (512 experts, 10 activated per token), it outperforms Gemini 2.5 Flash Thinking on several benchmarks, including AIME 2025 and Arena-Hard, while staying computationally efficient. Its strengths show clearly on competition math and demanding coding challenges. It is best suited to educational AI, math tutoring, code debugging systems, and applications where interpretable reasoning is important. The open-weight release allows fine-tuning for specialized reasoning domains and local deployment.

Open in playground Try in chat Compare

Key info

Input

Output

Features

Context window

128K

Max output

66K

Input price

$0.15 /1M

Output price

$1.20 /1M

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

Available routes

Qwen 3 Next 80B A3B Thinking runs on 2 different routes 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
Nebius	EU	Enterprise	No	$0.15	$1.20
Novita	US	Enterprise	No	$0.15	$1.50

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

Data handling per route

Each route hosting Qwen 3 Next 80B A3B Thinking has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

Nebius — Finland🇫🇮

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
Third-party access: None disclosed
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Novita — United States🇺🇸

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 Qwen 3 Next 80B A3B Thinking 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: "nebius/qwen3-next-80b-a3b-thinking",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);