Open Models Almanac
Methodsinference & servingdecoding strategy

general family · filed under inference & serving

Greedy decoding

Selects the highest-scoring next output without sampling; evidence includes autoregressive generation and action generation.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See decoding strategy for the methods filed beside it.

Also called Greedy action generation without sampling, do_sample=False, Greedy autoregressive decoding, simple greedy autoregressive decoding, Greedy decoding for benchmark generation, greedy decoding without reasoning.

sources
4
models
4
labs adopt it
4
strongest
default

Documented in

Physical Intelligence
Aleph Alpha
Liquid AI
OpenVLA project (Stanford et al.)

Evidence

4 quotes from 4 sources, one card per model, each opening with its strongest.

How sources treat it
used 3default 1

Kolibri-1

Aleph Alphaadopted1 source

For efficiency reasons, judge requests use greedy decoding without reasoning.

usedevaluation onlyin Kolibri RL training runAleph Alpha
↑