Alibaba's Qwen3.8 Flash Next: a 125B-parameter MoE language model with about 6B active parameters, previewing Qwen 4 design.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Qwen3.8-Flash-Next is a language model from Alibaba's Qwen team, announced on the Qwen blog. According to a YouTube creator's summary, it is a mixture-of-experts model with 125B total and about 6B active parameters. It is described as a preview of the architecture planned for Qwen 4.
The reported architecture combines gated DeltaNet layers, Qwen sparse attention and N-gram embeddings. The same source says it outperforms DeepSeek V4 Flash on benchmarks and is inexpensive to run. These claims come from a third-party video and have not been verified against the official page, whose text could not be retrieved.
The creator says the weights are open. The license, repository, Hugging Face page, API pricing and exact release date were not found in the material provided, so they are left unspecified here. The model would normally be reachable through Qwen's own chat and API products, and open weights would allow self-hosting.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
125B MoE with 6B active, a preview of the architecture for Qwen 4 using gated Delta Net layers, Qwen sparse attention and N-gram embeddings. It beats DeepSeek v4 Flash on benchmarks, is cheap, and weights are open.
Open-weight 125B-parameter model, larger than Qwen3.8 27B. It scores about 56 on Artificial Analysis and ranked 19th on Beautybench at about a penny and 26k tokens.
Closest neighbours by meaning, not just by category.
Model · Alibaba
Alibaba's open-weights 2.4T-parameter MoE language model (95B active) from the Qwen3.8 family, with a hosted Max version.
Model · Alibaba
Alibaba's Qwen 3.5 large open-weights language model (~122B parameters) from the Qwen family.
Model · Alibaba
Small 0.8B-parameter open-weights language model in Alibaba's Qwen 3.5 family, aimed at lightweight and on-device use.
Model · Alibaba
Compact 27B-parameter model from Alibaba's Qwen family, runnable locally on a single high-end consumer GPU.
Model · Alibaba
Small 0.5B-parameter open-weights base language model from Alibaba's Qwen2.5 series, Apache 2.0 licensed.
Model · Alibaba
Alibaba's fast multimodal Qwen model with text, image, audio and video input, 1M-token context and real-time streaming.
