Qwen-Flash
"The Qwen3 Flash model offers a powerful fusion of thinking and non-thinking modes with dynamic in-conversation switching, excelling in complex reasoning while showing significant gains in instruction following and text comprehension.".
text → text · made by Alibaba
$0.05→$0.4per Mtok in / out
Alibaba
Sold by 2 ways
| Seller | Lane | Rate |
|---|---|---|
| Alibaba api | in<=256k | $0.05per Mtok in$0.4per Mtok outalibabacloud.com · read 2026-08-24 |
| Alibaba api | in-256k-1m | $0.25per Mtok in$2per Mtok outalibabacloud.com · read 2026-08-24 |
About
Qwen-Flash — a text model from Alibaba.
It takes text and returns text, with a context window of 1,000,000 tokens. It was published in July 2025, trained on material up to April 2024. Its sellers say it can reason step by step and call a tool. The catalogue files it under chat.
Every current figure
- Maker
- Alibaba
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 1,000,000 tokens
- Longest answer
- 250,000 tokens
- Published
- July 2025
- Knowledge to
- April 2024
- Licence
- not read