Qwen3.8-Flash-Next
Weights availableOpen-weight Qwen model with a 256K-token context window, image input and tool calling. Released August 26, 2026. Weights on Hugging Face.
Access details: Qwen open weights (Hugging Face)
The Hugging Face repository. Download the weights and run them yourself or with a host.
Qwen/Qwen3.8-Flash-NextUpdated Oct 1, 2026
Capabilities & limits
- Context window
- 256K
- Maximum output
- Not published
- Image input
- Supported
- Tool calling
- Supported
- Structured output
- Not published
Hosted price
USD / 1M tokens- Input
- $0.15
- Output
- $0.47
What OpenRouter charges for qwen/qwen3.8-flash. Other hosts set their own prices.
Free to download. Running the weights costs whatever your hardware or host charges.
Price history
Oct 1, 2026
Input Not applicable per million tokens
Oct 1, 2026
Output Not applicable per million tokens
Technical details
- Parameters
- 125B with 6B activated, plus 51B n-gram embedding and 4B MTP
- Active parameters
- 6B
- Architecture
- Causal Language Model with Vision Encoder
- Training / knowledge cutoff
- Not published
- Input modalities
- Text, image, video
- Output modalities
- Text
License
Other: qwen-community-1.0 (as declared in the Hugging Face model card metadata)
From Qwen
Official announcement ↗How they describe it
This experimental preview of the architecture that will underpin Qwen4
Release & lifecycle
- Published
- Aug 26, 2026
- available
- As of Oct 1, 2026