Pathway's 150M-parameter BDH reasons in latent space, beyond Transformers
bigdata · x · 2026-09-27
Pathway CEO Zuzanna Stamirowska argues transformers aren't enough for true reasoning. Their BDH model (150M params) reasons in latent space, achieves ARC-AGI results at dramatically lower inference cost, learns continually via fast weights, and maintains state over long horizons. She contends long-horizon agents may require a fundamentally different architecture, targeting higher intelligence per watt via enterprise one-rack deployments, open weights and on-device AI.
More from Infra
- India's Sarvam launches Token Factory, hosting top open models on local infra — HarveenChadha · 2026-11-16
- Gemini 3.5 Flash now runs on air-gapped Google Distributed Cloud, offline up to 12 months — shashib · 2026-11-16
- Gemma.c: full Gemma 1/2/3 multimodal inference in a single C file, ~72 tok/s on CPU at INT8 — This-Purchase-3325 · 2026-11-16
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Jailbroken PS4 runs a 9B open-source model at up to 43 tok/s by talking directly to its 2013-era GPU — AIFlow_ML · 2026-11-16
- Mobile AI Factory: agents auto-build on-device SLMs with 19x less GPU time — pmttyji · 2026-11-16