A decentralized system that automates ML research and trains domain-expert 1.58-bit ternary models for CPU-native inference.
March 30, 2026
Original Paper
MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training
arXiv · 2603.25813
The Takeaway
It combines autonomous hyperparameter/data iteration with BitNet b1.58, allowing high-performance expert models to be researched, trained, and served on commodity CPU hardware without GPUs.
From the abstract
We present MAGNET (Model Autonomously Growing Network), a decentralized system for autonomous generation, training, and serving of domain-expert language models across commodity hardware. MAGNET integrates four components: (1) autoresearch, an autonomous ML research pipeline that automates dataset generation, hyperparameter exploration, evaluation, and error-driven iteration; (2) BitNet b1.58 ternary training, enabling CPU-native inference viathis http URLwithout GPU hardware; (3) DiLoCo-based d