Qwen3.8-Flash-Next
An experimental open-weight model from Alibaba's Qwen team, released in August 2026 as a preview of the architecture behind Qwen4.
Qwen3.8-Flash-Next is an open-weight model released by Alibaba’s Qwen lab on 26 August 2026, with the full weights and an FP8 version published on Hugging Face and ModelScope. It has roughly 125 billion total parameters and a context window of about 262,000 tokens, meaning it can hold a very large amount of text in mind at once.
The interesting part is the label. Qwen presented it as an experimental preview of the architecture that will underpin Qwen4, so it works as an early look at where the family is heading rather than as a finished flagship. Local AI tools picked it up quickly, with KoboldCpp adding support three days after release, though quantised copies of varying quality appeared just as fast.