TLDRocket
Sign in

Alibaba releases Qwen3.8-Flash-Next, an open-weight multimodal Mixture-of-Experts model

Open source release ● Confirmed 86% confidence first seen

Alibaba’s Qwen team released Qwen3.8-Flash-Next, an open-weights multimodal Mixture-of-Experts model designed for low cost per token. The model is described as an early preview of the architecture intended for Qwen4 and uses a structure with a smaller active parameter set per token.

Alibaba GLM-5.3-Flash Qwen Qwen3.8-Flash Qwen3.8-Flash-Next Z.ai DGX Spark Gated DeltaNet MarkTechPost MiniMax Muon Muon training Qwen 3.8 Max Qwen Sparse Attention Qwen3.7-Plus Qwen4 The New Stack UD-IQ1_S UD-Q2_K_XL Unsloth

Source coverage

Alibaba’s Qwen Team Releases Qwen3.8-Flash-Next: A 125B Multimodal MoE With 6B Active Parameters Previewing the Qwen4 Architecture MarkTechPost Qwen3.8-Flash-Next Simon Willison’s Weblog industry analysis Alibaba just released Qwen3.8-Flash: “An early preview of the architecture in Qwen4” The New Stack GLM-5.3-Flash vs Qwen3.8-Flash-Next: Two Chinese AI Labs Independently Converge on the Same Model Architecture MarkTechPost

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.