aiqbench analysis
Alibaba's Qwen3.8-Flash-Next introduces a significant architectural shift by combining Gated DeltaNet and Sparse Attention to drastically reduce active parameters per token while maintaining multimodal capabilities. This design achieves roughly one-ninth of the training cost of its predecessor, though deploying the 172 GiB FP8 checkpoint requires substantial infrastructure like GB300 clusters rather than standard workstations. The release serves as a critical preview for the upcoming Qwen4 architecture, highlighting how efficiency gains are being prioritized alongside model scale.
Alibaba’s Qwen Team Releases Qwen3.8-Flash-Next: A 125B Multimodal MoE With 6B Active Parameters Previewing the Qwen4 Architecture
We look at Qwen3.8-Flash-Next, Alibaba's open-weight multimodal Mixture-of-Experts model and an early preview of the Qwen4 architecture. We break down where the 180B parameters actually sit: a 125B backbone, a 51B N-gram embedding table, and a 4B multi-token prediction module, with only 6B active per token. We walk through the four architectural changes — the Gated DeltaNet and Qwen Sparse Attention hybrid, Gated Residual, N-gram Embedding, and the Muon optimizer. We also cover the benchmark results, the reported 1/9 training cost against Qwen3.7-Plus, and what self-hosting a 172.78 GiB FP8 checkpoint really demands. The post Alibaba’s Qwen Team Releases Qwen3.8-Flash-Next: A 125B Multimodal MoE With 6B Active Parameters Previewing the Qwen4 Architecture appeared first on MarkTechPost.
Alibaba’s Qwen team has released Qwen3.8-Flash-Next, an open-weight multimodal Mixture-of-Experts model built for cost per token. The checkpoint pairs a 125B backbone with a 51B N-gram embedding table and a 4B multi-token prediction module. Only 6B parameters activate per token. The team positions it as an early preview of the architecture that will underpin Qwen4, the same role Qwen3-Next played for Qwen3.5. Four changes carry the release: a Gated DeltaNet and…
Excerpt shown under the publisher's RSS terms · 71 words · full text and figures are on the publisher's site
Related coverage
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
GLM-5.3-Flash vs Qwen3.8-Flash-Next: Two Chinese AI Labs Independently Converge on the Same Model Architecture
MarkTechPost · Aug 28, 2026
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · Sep 17, 2026
Alibaba’s Tongyi Lab Releases Qwen-Audio-3.0-TTS, a Hosted Text-to-Speech Model in Flash and Plus Tiers Across 16 Languages
MarkTechPost · Jul 20, 2026