Skip to main content

Self-hosted preview of Retool Agents

Learn how to install a self-hosted preview of Retool Agents.

Version: Cloud and 3.227 Edge

A preview release of Retool Agents is currently available to self-hosted organizations for deployment into a non-production environment. This preview release is intended to allow customers to preview and create proof-of-concepts that show how agents may work in their environment.

As Retool is continuously working to develop and improve Retool Agents to meet the high standards required for production environments, this preview release is not intended for production use and this build should not be deployed in an existing Retool installation or deployment. Retool recommends that self-hosted organizations create a new deployment in a private network alongside resources to which you intend to connect. Retool reserves the right to make changes to the agents preview without notification.

This preview is available at the agents-preview tag of the tryretool/backend Docker image.

This preview release includes some infrastructure differences and sends limited agents-specific data to Retool’s cloud. Find more details in the FAQ.

Requirements​

To use the preview release, your organization must:

  • Be on a Team, Business, or Enterprise plan.
  • Have its own AI model key(s).
  • Meet all of the requirements outlined in the tutorial for your chosen deployment infrastructure.

Limitations​

The agents preview release does not currently support deploying on:

Installation​

The agents preview release is supported for the following self-hosted deployment options.

The process to deploy the agents preview release with Docker Compose is slightly different than a traditional deployment of Self-hosted Retool.

Follow the instructions outlined in the tutorial corresponding to your deployment infrastructure:

The following step diverges from the standard process:

When cloning the tryretool/retool-onpremise git repository, pull the agents-preview branch.

git clone -b agents-preview https://github.com/tryretool/retool-onpremise

Post-installation​

Once you’ve logged in, navigate to <your domain>/resources/retool_ai and configure a model provider.

You must provide your own API key for OpenAI, Anthropic, Azure OpenAI, or Google to use Retool Agents. For more information, refer to the Retool AI providers and models page.

Frequently Asked Questions​

This FAQ aims to provide answers to common questions about the self-hosted preview release of Retool Agents. For deployment-agnostic questions, refer to the Agents FAQ.

How is the preview release billed?​

During the preview release period, you will not be billed for agents usage. You will be billed as normal for all other usage.

I’ve hit a free usage limit. How do I continue to use agents?​

Retool organizations receive up to $50 worth of agent run time per month. Because Self-hosted agents deployments of agents require you to use a self-managed LLM, which costs $5/hour, you receive get 10 hours of free usage per month. Refer to Model pricing for more information.

If you've exceeded the cap, you'll see a warning message pop up when you attempt to send a message to an agent, and the agent will not be executed. Reach out to your account team for help getting unblocked.

What are the notable differences between a preview release deployment and a standard deployment of self-hosted Retool?​

The infrastructure of the agents preview deployment is very similar to a Retool deployment with workflows. There are a few container differences that you can find by viewing a diff of the agents-preview branch in the retool-onpremise repo.

It is important to note that the use of a Retool-managed agents observability API is required in the preview release. This API strictly does not receive, persist, or have access to sensitive customer data, agent inputs, or agent outputs. This API is required to power the Monitor dashboard.

What data is sent to the Retool-managed agents observability API?​

The data sent to the Retool-managed (cloud-hosted) agents observability API consists of the following:

  • UUIDs corresponding to objects in the Retool deployment’s Postgres database.
  • Success/failure status of agent runs.
  • Model type of agent runs.
  • Runtime of agent runs.
  • Number of LLM tokens used by agent runs.
  • Information about the tools used during agent runs. Tool information is limited to UUIDs and the configured tool name and tool parameter names.

No customer data beyond the names of configured tools and tool parameters are sent to this API.

If you have any feedback or concerns about this data, please reach out to your account team.

Does the preview release send any agent inputs and outputs to Retool servers?​

No agent inputs or outputs are sent back to Retool servers.

When will agents be included in an edge or stable release?​

Agents will be available in public beta to all self-hosted organizations in an upcoming edge release and in the Q3 stable release.