Open Models Almanac
Methodsinference & servingreasoning control

implementation detail · filed under inference & serving

clear_thinking chat-template flag

The clear_thinking flag controls a chat-template behavior and is reported to default to false when omitted.

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

Also called clear_thinking.

sources
2
models
2
lab adopt it
1
strongest
default

Documented in

Z.ai (Zhipu AI)

Evidence

2 quotes from 2 sources, one card per model, each opening with its strongest.

How sources treat it
optional 1default 1

GLM-5.3

Z.ai (Zhipu AI)adopted1 source

In the chat template for GLM-5.3, clear_thinking defaults to false if not passed. For chat scenarios, explicitly pass clear_thinking=true.

defaultunclearin GLM-5.3Z.ai
↑