Open-weight models compress model-layer margins but boost AI infra demand, argues Baker
GavinSBaker · x · 2026-10-09
Investor Gavin Baker argues open-weight models are net positive for AI infrastructure demand despite compressing margins at the model layer: an open-weight token consumes roughly the same compute as a frontier token for a similar-size model, so cheaper models grow inference demand rather than shrink it.
Related event: Open-Weight Models Squeeze Model-Layer Profits but Boost Compute Demand(2 posts)→
More from Venture
- SoftBank to acquire RAI Institute, the robotics AI lab founded by Boston Dynamics creator Marc Raibert — lukas_m_ziegler · 2026-10-09
- Oxide Computer announces $445M Series D — ahlCVA · 2026-10-09
- Jevons Effects Keep Open and Closed AI Models Reinforcing Each Other — soumitrashukla9 · 2026-10-09
- Node.js Creator Ryan Dahl's Deno Team Is Joining Cloudflare, Merging with workerd — ilreb · 2026-10-09
- Node.js Creator Ryan Dahl's Deno Team Is Joining Cloudflare, Merging with workerd — m-app · 2026-10-09
- Dev's AI-built SEO micro-SaaS portfolio hits $1k MRR with barely any maintenance — iannuttall · 2026-10-09