Qwen 3.8 Flash Next releasing: 125B params with 51B N-gram
zephyr_z9 · x · 2026-08-25
Qwen 3.8 Flash Next is set to release tomorrow. The model features 125 billion parameters, including 51 billion N-gram parameters and 6 billion active parameters, based on the next-generation Qwen 4 architecture.
Related event: Alibaba Announces Open Source Qwen3.8-Flash-Next(4 posts)→
More from Models
- Test of 1,000 prompts: ChatGPT and Google AI share only 2.6% of cited sources — nikvassev · 2026-08-25
- Rumor: Alibaba's Qwen to release a 125B-A6B MoE model, hailing the era of ~120B models — TheZachMueller · 2026-08-25
- Unsloth AI aims for day-zero llama.cpp support for Qwen models — danielhanchen · 2026-08-25
- MiniMax Launches H3 Model and End-to-End Design Platform — mhdfaran · 2026-08-25
- Qwen Announces Qwen3.8-Flash-Next Open Source Release — cedric_chee · 2026-08-25
- AI detection tool flags content as likely generated by Claude — pedrodias · 2026-08-25