Qwen3.8-Flash-Next
Open in the interactive archive →
Overview
Preview of Alibaba's next-generation efficiency model: 125B total / 6B active MoE on an early Qwen4 architecture; very low per-token pricing.
Generates
App available
Details
- Vendor
- Alibaba
- Type
- Multimodal LLM
- Release year
- 2026
- Jurisdiction
- China
- CEO
- Eddie Wu
- Key staffers
- Jingren Zhou (CTO, Tongyi Lab)
- Deployment
- Cloud (Alibaba Cloud API; Qwen4 architecture preview)
- Open source
- No
- Free tier
- Yes
- Subscription
- API: $0.15/M input, $0.47/M output.
- Stock status
- $145 (BABA, May 2026)
- X
- @AlibabaGroup
Known safety issues
None publicly documented (preview)
AIDb Rating
AIDb v2.0 (curated): Capability 7/10 · Currency 10/10 · Access 7/10 · Significance 6/10 → 7.50
Last verified 2026-09-14 · Primary source ↗
Cite this
“Qwen3.8-Flash-Next.” AIDb — The AI Database. Verified 2026-09-14. https://theaidatabase.com/m/alibaba-qwen3-8-flash-next.html
More from Alibaba
Babel-83B · GTE-Large · ModelScope T2V · ParScale · PolyLM · Qwen · Qwen 1.5 · Qwen-72B