Model techniques map
Techniques

general family · not yet filed

Pre-training methods

Also called New pre-training methods.

source
1
model
1
labs adopt it
0
strongest
mentioned

How sources treat it

One count per evidence span, weakest treatment to strongest.

mentioned 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

New pre-training methods + larger-scale RL post-training deliver benchmark results ahead of flagship models

mentionedotherin DeepSeek-V4.1-FlashDeepSeek