Greenblatt: 3x jump in model depth points to architecture changes, not scaling

RyanGreenblatt · x · 2026-09-04

Pushing back on a claim that a new model's reasoning depth is "within a factor of 2 of GPT-4," Ryan Greenblatt argues the gain is mostly architecture-driven: open-weight scaling trends suggest a 3x depth increase would normally require 81x more parameters (equivalent to >6000x more compute), so something beyond normal scaling must be at play.

Related event: Greenblatt: 3x Depth Gain Comes From Architecture, Not Parameter Scaling(4 posts)→

Original post →

More from Models

Models channel →