DevMachine Learning25 min reading time

[AINews] Qwen 3.8 Max(2.4T) and 27B, new open weights models for Coding and Cowork

Latent.Space
Read full post
Alibaba's Qwen lab is releasing open weights for Qwen 3.8 Max, a 2.4 trillion parameter model excelling in autonomous coding, AI research, hardware design, and multimodal tasks. The model demonstrates advanced capabilities including long-horizon autonomous coding, outperforming humans in data science challenges, and integrating visual feedback for multimodal agents.

More on this story


More in Dev

Introducing the Agents API

Covered by 3 sources
Dev1 min read

Datasette 1.0a39 and 0.65.4 security releases

Simon Willison's Weblog
Dev1 min read

Native is now the future of mobile at Shopify

Simon Willison's Weblog