AnnouncedAug 26, 2026
Qwen Teases 3.8-Flash-Next, a 125B Sparse MoE
Alibaba's next Qwen release pairs a large parameter pool with a tiny active footprint, promising speed without the full compute bill.
Qwen · Alibaba·Text / LLM
On the horizon
Announced and rumored open models, with expected timing. Each is auto-promoted into the feed the moment it actually ships.
Alibaba's next Qwen release pairs a large parameter pool with a tiny active footprint, promising speed without the full compute bill.
The new mixture-of-experts model is billed as more capable than V4 Pro while costing less to run, and ships under an MIT license.
The Chinese AI lab says its stealth-tested system belongs to the GLM series and will ship with open weights under an MIT license.
An experimental open-weight model swaps token-by-token decoding for discrete diffusion, generating text in parallel blocks.