factual

Test-time and training compute are tradeable

Andy Jones's 'Scaling Scaling Laws with Board Games' showed you can trade off test-time compute and training compute—spending more compute on MCTS search yields performance equivalent to more training compute—anticipating LLM inference scaling, and additionally predicted how much compute is needed to solve a larger version of a board game.

factualpending

Speaker

Dwarkesh Patel

Evidence Quote

“he basically anticipated inference compute, or inference scaling, by showing that you can trade off test-time compute and training compute”

Source

Dwarkesh Patel Podcast - Eric Jang – Building AlphaGo from scratch— Dwarkesh Patel Podcast
Created: 6/13/2026, 3:37:56 AM

Your notes

Loading notes...