Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod
Pathway's brain-inspired post-transformer model reasons in latent space, not tokens—and hits a new cost-efficiency record on ARC-AGI.
Why it matters
A novel reasoning architecture (latent-space reasoning vs. chain-of-thought) is being scaled on AWS infrastructure and benchmarking competitively. This matters to practitioners exploring alternatives to standard transformer token-emission patterns and to enthusiasts tracking the lab race for post-transformer approaches.
The key facts
9 to knowPathway's Baby Dragon Hatchling (BDH): post-transformer architecture with latent-space reasoning
BDH reasons in latent space instead of emitting chain-of-thought tokens
BDH-CQ achieved new cost-efficiency record on ARC-AGI-1 benchmark
Development scaled on Amazon SageMaker HyperPod
Published September 8, 2026
Pathway's Baby Dragon Hatchling (BDH) is a brain-inspired, post-transformer architecture
BDH-CQ set a new cost-efficiency mark on ARC-AGI-1 benchmark
Developed and scaled on Amazon SageMaker HyperPod
Published Sep 2026 on AWS machine-learning blog
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: Pathway's Baby Dragon Hatchling (BDH) is a brain-inspired, post-transformer architecture that reasons in latent space instead of emitting chain-of-thought tokens. See how Pathway develops and scales BDH on Amazon SageMaker HyperPod, and how BDH-CQ set a new cost-efficiency mark on the ARC-AGI-1…