When was Qwen3.8 Flash-Next released?

Alibaba released Qwen3.8 Flash-Next on 26 Aug 2026. It is available through the qynio gateway.

Does Qwen3.8 Flash-Next support zero data retention?

Yes. Qwen3.8 Flash-Next is available with zero data retention by default on Sference, TensorX through the qynio gateway.

Where is Qwen3.8 Flash-Next hosted?

Qwen3.8 Flash-Next runs on crypto rails routes through the qynio gateway, with contractually enforced residency on supported providers.

Is Qwen3.8 Flash-Next trained on prompts and outputs?

No. Every hosting route for Qwen3.8 Flash-Next is configured so customer prompts and outputs are not used to train models by default.

Is a no-KYC deposit available for Qwen3.8 Flash-Next?

Yes. A no-KYC Data Processing Agreement is available on at least one route serving Qwen3.8 Flash-Next. Contact hello@qyn.io to receive it.

Qwen3.8 Flash-Next

by Alibaba

Released 12 days after Qwen3.8 27B · All Alibaba releases

Qwen3.8 Flash-Next is a natively multimodal MoE model from Alibaba with 125B total / 6B active parameters and a 256K context window. It targets frontier-grade coding and agentic work at very low active-parameter cost (Artificial Analysis intelligence index 55.8, coding index 73.1).

Open in playground Try in chat Compare

Key info

Input

Output

Features

Context window

262K

Max output

—

Input price

$0.20 /1M

Output price

$0.50 /1M

Released

26 Aug 2026

onchain settlement available
Zero data retention on pay-as-you-go
No training by default
no-KYC deposit available

Available routes

Qwen3.8 Flash-Next runs on 2 different routes through the qynio gateway. Compare residency, zero retention, and training posture at a glance — full data-handling detail per route below.

Provider	Region	Zero data retention	Training	Input	Output
Sference	EU	Zero data retention	No	$0.20	$0.50
TensorX	EU	Zero data retention	No	$0.20	$0.50

Data handling per route

Each route hosting Qwen3.8 Flash-Next has its own privacy posture, residency, and no-KYC terms. Postures are maintained by qynio with a last-verification timestamp.

Sference — United Kingdom🇬🇧

Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. EEA; wallets; deposit available.

Zero data retention: On by default on Pay-as-you-go. Derived from the logging and moderation facts.
Training: No training on customer data.
Logging: None
Moderation: Not established
Caching: Not established
Subprocessor access: No subprocessor reads content
no-KYC deposit: deposit available
Transfer mechanism: wallets

TensorX — European Union🇪🇺

Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. EU; deposit available.

Zero data retention: On by default on Pay-as-you-go. Derived from the logging and moderation facts.
Training: No training on customer data.
Logging: None
Moderation: Not established
Caching: Not established
Subprocessor access: Not established
no-KYC deposit: deposit 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#36 of 630 LLMs

TierFrontier

Output speed86 tok/s

First token1.33s

Intelligence Index55.8

Coding Index73.1

Reasoning & knowledge

GPQA Diamond

92%

Humanity's Last Exam

38%

Long-context reasoning

77%

Coding

SciCode

47%

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

Get started

Call Qwen3.8 Flash-Next through the qynio gateway with one API key. Let your coding agent set it up, or call it directly — qynio is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.

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 qynio for you.

Or call it directly

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.QYN_API_KEY,
5  baseURL: "https://api.qyn.io/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "sference/Qwen/Qwen3.8-Flash-Next",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);