When creating a new REST API resource, you can now connect an OpenAPI or Swagger specification. Connecting an API specification enables:

• Autocomplete: Get endpoint and parameter suggestions when writing queries.
• Validation: Retool validates requests against the specification schema.
• Better Assist: Assist can more effectively generate queries based on the specification.
• Documentation: Endpoint descriptions from the specification appear in the query editor.

When working with resources, users can also manually create endpoints or parameters that aren't included in the specification.

Starting February 23rd, Retool will roll out provider-specific AI resource integrations (e.g., OpenAI, Anthropic, and Bedrock) in place of the singular Retool AI resource. This change is part of a broader effort to make AI integrations behave more like other resources in Retool: more flexible, more governable, and easier to scale as usage grows.

These changes will first roll out to cloud instances. Self-hosted instances can expect to receive the changes in a future release. There are no pricing or packaging changes associated with this update.

No action is required to facilitate this change. All existing apps, queries, workflows, and agents will continue to work based on existing configurations.

What's changing​

Instead of a single Retool AI resource, AI providers will function like any other resource at Retool. This allows for more flexibility when using AI, such as:

• Using multiple API keys per provider (e.g., by team, resource environment, or use case) so you can better manage access and usage limits.
• The same granular permissions and governance controls, allowing admins to decide who can create or modify specific AI resources.
• Clearer auditability and visibility into which apps and workflows are using which provider and key.
• Easier multi-provider setups, letting you mix and match models from different providers in the same workspace.

The dedicated AI Action query is also being deprecated as its functionality will be available in AI resource queries.

Retool released a Version 3.0 of the Microsoft SQL Server integration that includes various fixes and improvements. New Microsoft SQL resources will use this version by default. Existing resources can be updated using the Connector version setting.

Beginning in the 2026 Q3 stable release, Version 1.0 and Version 2.0 of the integration will be deprecated. Retool strongly recommends that you update your resource configuration before Q3 in order to prevent any interruptions and to resolve any unforeseen issues. If necessary, Retool may automatically migrate resources to Version 3.0 in order to ensure any issues are identified and resolved prior to the removal of the old versions.

There is a known breaking change introduced in Version 2.0 that will also apply when upgrading from Version 1.0 to Version 3.0:

Beginning in Version 2.0, array inputs must be passed as a string when used in string_split in order to pass the Microsoft SQL SDK validation. For example, select * from string_split({{[1,2,3]}}, ','); must change to select * from string_split({{[1,2,3].join(',')}}, ',');.

No breaking changes are expected between Version 2.0 and Version 3.0.

Self-hosted organizations can now restrict the creation or usage of specific resource types using either the RESOURCE_TYPES_CREATION_DENY_LIST or RESOURCE_TYPES_DENY_LIST environment variables.

• RESOURCE_TYPES_CREATION_DENY_LIST prevents users from creating any resources of the specified types. Any existing resources continue to function.
• RESOURCE_TYPES_DENY_LIST prevents users from both creating and using any resources of the specified types. Any related resource queries will then fail.

Any restrictions you set can be reverted at any time.

Retool removed some JDBC connectors that were inadvertently included in certain self-hosted release versions. These include:

• Actian Ingres/Vector JDBC Driver.
• Clickhouse JDBC Driver.
• IBM Data Server Driver for JDBC.
• IBM Informix JDBC Driver.
• Trino JDBC driver.

These JDBC connectors are not supported by Retool for self-hosted deployments. If you have created a JDBC resource using these connectors, update its configuration as needed to make use of your own preferred JDBC drivers.

Retool has since released patch updates for affected releases that remove these connectors.

Retool has deprecated version 1.0 of the Databricks connector. If your organization has any Databricks resources that still use this connector version, update the resource configuration to use the version 2.0 connector.

Self-hosted customers must also enable the Java DBConnector to use version 2.0 of Databricks connector.

Change the connector version from the Databricks resource's configuration settings. Any existing queries that previously worked with version 1.0 will continue to work with version 2.0.

The option to use version 1.0 of the Databricks connector will be removed in Q2 2025 from cloud instances and subsequent releases for self-hosted instances.

Retool has deprecated version 1.0 of the MySQL connector. If your organization has any MySQL resources that still use this connector version, update their configuration to use the version 2.0 connector.

The option to use version 1.0 of the MySQL connector will be removed in Q2 2025 from cloud instances and subsequent releases of self-hosted Retool.

Retool will deprecate the use of YAML for protected apps, queries, and resources in Source Control. Support will be removed in self-hosted Retool 3.196 and later.

Self-hosted organizations still using YAML serialization of these elements in Source Control must migrate to Toolscript prior to upgrading.

Retool now supports Amazon SNS. You can use the Amazon SNS integration in apps and workflows to publish messages to a topic. Once configured, you can subscribe to the topic using Amazon SQS, enabling you to use Retool's SQS integration to respond to events as they occur.

Retool now supports Amazon SQS. You can use the Amazon SQS integration in Retool Apps and Retool Workflows to send, receive, and delete messages from a queue. This integration enables you to respond to events as they occur, triggering changes and actions in your workflows or apps.