Open Models Almanac
Methodsinference & servingagentic scaffolding

general family · filed under inference & serving

Advisor strategy

Advisor Mode is an implementation of the advisor strategy where the model runs the agentic loop end-to-end.1 The model escalates to a larger advisor model only at specific inflection points, such as planning or recovering from repeated failures.2

Trade-off The method keeps most of the run at executor cost while maintaining high performance.3

How this definition was written

Summarised by glm-5.3-flash from 2 sources; every sentence is checked against the quote it cites.

  1. Step 3.7 Flash supports Advisor Mode, StepFun’s implementation of the advisor strategy described by Anthropic — StepFun Releases Step 3.7 Flash: A 198B MoE Vision-Language Model for Coding Agents and Search Workflows
  2. The model runs the agentic loop end-to-end — calling tools, reading results, iterating — and escalates to a larger advisor model only at specific inflection points, such as planning or recovering from repeated failures — StepFun Releases Step 3.7 Flash: A 198B MoE Vision-Language Model for Coding Agents and Search Workflows
  3. keeping most of the run at executor cost — Step 3.7 Flash

Also called Advisor Mode.

sources
2
model
1
lab adopt it
1
strongest
optional

Documented in

Evidence

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

How sources treat it
mentioned 2optional 2

Step-3.7-Flash

StepFunadopted2 sources

To push quality further without giving up Flash-tier efficiency, Step 3.7 Flash supports Advisor Mode.

optionalinference servingin Step 3.7 FlashStepFun
Step 3.7 Flash·StepFun·2026-05-09
3 more from this model

This is Step's implementation of the advisor strategy described by Anthropic

mentionedinference servingin Step 3.7 FlashStepFun
Step 3.7 Flash·StepFun·2026-05-09

Step 3.7 Flash supports Advisor Mode, StepFun’s implementation of the advisor strategy described by Anthropic

optionalinference servingin Step 3.7 FlashStepFun

StepFun’s implementation of the advisor strategy described by Anthropic

mentionedinference servingin Step 3.7 FlashStepFun
↑