Skip to main content

Documentation Index

Fetch the complete documentation index at: https://ai-kb.automationanywhere.com/llms.txt

Use this file to discover all available pages before exploring further.

Effective date: June 1st, 2026 · Previous default: GPT-4.1 mini · Action required: None, unless opting out

​
Overview

ChangeDefault system model update
New defaultGPT-5.4 nano
Previous defaultGPT-4.1 mini
Effective dateJune 1st, 2026
Action requiredNone, unless opting out
On June 1st, 2026, GPT-5.4 nano becomes the default system model in Enterprise Knowledge. This change provides a more capable, cost-efficient, and future-ready foundation for all AI-powered workflows and agents.

​
What’s improving

Newer model generation

GPT-5.4 nano is part of the latest model family, designed for stronger reasoning and more reliable outputs.

Increased output capacity

Supports significantly larger responses, enabling more detailed workflows, plans, and results.

Improved tool usage and reasoning

Better at multi-step thinking, tool selection, and orchestrating complex agent tasks.

More recent training data

Updated knowledge base leads to better accuracy across modern use cases.

​
Cost impact

GPT-5.4 nano is more cost-efficient for most workloads. Some tasks may generate longer outputs due to improved reasoning, but the overall expected impact is equal or lower cost per execution.
FactorImpact
Cost per tokenLower than GPT-4.1 mini
Output lengthMay increase for complex tasks
Expected net costEqual or lower per execution

​
Important: opting out

If you have workflows or agents that rely on GPT-4.1 mini and do not want them to migrate to the new default, you need to explicitly configure them before June 1st, 2026.
If no action is taken before June 1st, 2026, all agents and workflows using the system default will automatically switch to GPT-5.4 nano.
1

Create a dedicated agent

Set up an agent specifically for the workflows that need to retain the previous model.
2

Set the model to GPT-4.1 mini

In the agent’s General tab, set the model to GPT-4.1 mini explicitly. This overrides the system default and ensures consistent behavior for existing implementations.

​
Model configuration options

You can control model usage at multiple levels:
  • Agent level — Set a specific model in the agent’s General tab. This overrides the system default for all workflows using that agent.
  • Workflow level — Override the model for a specific workflow where required.
For instructions on setting a model at the agent level, see Predefined and Custom Models.

​
Summary

What's better

  • Better intelligence
  • Stronger agent capabilities
  • Larger outputs
  • Lower cost

What you need to do

Nothing — unless you want to keep using GPT-4.1 mini. In that case, set the model explicitly at the agent level before June 1st, 2026.

​
Support

For questions about this change or help configuring your agents, contact your Automation Anywhere account manager or open a support ticket through the Automation Anywhere support portal.

Default Model Changelog

View the full history of default model changes in Enterprise Knowledge.