Model techniques map
Techniquesoptimizationoptimizer

implementation detail · filed under optimization

Eight-step Newton–Schulz iteration

Using eight Newton–Schulz steps for Muon orthogonalization, selected for added stability under stress.

Also called Eight-step Newton–Schulz orthogonalization, Newton–Schulz iteration to 8 steps.

source
1
model
1
lab adopt it
1
strongest
default

How sources treat it

One count per evidence span, weakest treatment to strongest.

default 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

We set the Newton–Schulz iteration to 8 steps, favouring the additional stability under stress.

defaultunclearin Qwen3.8-Flash-NextQwen

Filed alongside

Other methods under optimization :: optimizer.