Pathway's 150M-parameter BDH-CQ scores 29.5% on ARC-AGI-1 at $0.0007 per task.

Built on a post-transformer BDH architecture that reasons recurrently in latent space, it hit 29.5% pass@2 on public ARC-AGI-1 at a computed cost roughly 11x cheaper per task than GPT-5.6 Luna Low, even after OpenAI's 80% price cut on July 30 (Pathway). Be precise about what this claims: 29.5% is well below frontier accuracy. The claim is intelligence-per-dollar, and the frontier being pushed is the cost-accuracy Pareto curve, not the accuracy ceiling. Still interesting. 150 million parameters.