Open Models Almanac
Methodsinference & servingreasoning control

specific method · filed under inference & serving

Chain-of-thought reasoning

Chain-of-thought reasoning is a method that reasoning models use to work through complex problems.1 It allows the model to work through complex problems before providing an answer.1

How this definition was written

Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.

  1. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer — MiMo-V2.6-Pro - Intelligence, Performance & Price Analysis | Artificial Analysis

Also called Chain-of-Thought, COT, deep chain-of-thought, reasoning tokens, Chain-of-thought reasoning mode.

sources
6
models
5
labs adopt it
4
strongest
default

Documented in

Evidence

6 quotes from 6 sources, one card per model, each opening with its strongest.

How sources treat it
optional 2used 2default 2

Hy3

Tencent (Hunyuan)adopted2 sources

reasoning_effort: 'no_think' (default, direct response), 'low', 'high' (deep chain-of-thought)

optionalinference servingin Hy3Tencent Hy Team
1 more from this model

reasoning_effort: "no_think" (default), "low", "high" (deep chain-of-thought)

optionalinference servingin Hy3Tencent

Further reading

Picked by hand, not extracted: where to read more, not evidence for anything on this page.

↑