Morning Edition · Thursday, October 29, 2026Published at 2:14 AM EDT · New York
The compiled model, Qwenputer, runs C and MicroPython inside its forward pass and scores 97.5% to 100% on DeepMind arithmetic tasks, at the cost of doubling the checkpoint to 18.8 GB.

Divyansh Agrawal, an independent developer, published Qwenputer, a modified Qwen3.5-4B with a 32-bit RISC-V CPU built directly into its weights. The CPU consists of 11 transformer layers placed in front of Qwen's own layers. None of their w…
Track frontier labs, chips, export controls, model releases, regulation, and AI infrastructure.
The Global Intelligence Brief stays free.
Part of a tracked trend
Hybrid Linear-Attention Architectures
Open models will keep replacing much of full attention with fixed-size recurrent layers such as Gated DeltaNet, and that fixed state will increasingly be used as an engineered resource for memory, tool execution and cheaper long-context inference.
Start a discussion in Townsquare.
More from this edition
Comments
0No comments yet.