QwenLaunchAI ModelsAugust 9, 2026

Qwen releases Qwen3.8-2.4T-A95B, its largest open-weight MoE

Qwen3.8-2.4T-A95B is a sparse MoE with 2.4T total parameters and 95B active, supporting a 256K context window and multimodal input. Announced July 19, 2026 at the World AI Conference in Shanghai, it is now available on Together AI.

How this story unfolded

1 day · 1 report · 3 community posts · 4 of 8 shown

  1. Aug 12
  2. Aug 14

More stories today

Open the live feed