2026.0

AI Enterprise

• New models GPT-5.1 Codex Mini, GPT-5.1 Codex Max, GPT-5.2 Codex are now supported for OpenAI, Azure and OpenAI Compatible providers.

• Codex agent metrics are now included in the AI activity analytics.

• Models updated across AI providers:

  OpenAI Compatible

  New base models	Deprecated models
  Claude 4.1 Opus Claude 4.5 Opus Claude 4.6 Opus Claude 4.6 Sonnet Gemini 3.1 Pro	Claude 3.0 Haiku Claude 3.5 Haiku Claude 3.5 Sonnet Claude 3.7 Sonnet

  Amazon Bedrock

  New models	Deprecated models
  Claude 4.1 Opus Claude 4.5 Opus Claude 4.6 Opus Claude 4.6 Sonnet	Claude 3.0 Haiku Claude 3.5 Haiku Claude 3.5 Sonnet Claude 3.7 Sonnet

  Google Vertex

  New models	Deprecated models
  Gemini 3.1 Pro

• OpenAI and Azure OpenAI

  There are no longer any mandatory models required. The recommended models are GPT-5.2 and GPT-4o-mini

Web UI

Now you can clone profiles to expedite access management.

Code With Me

Code With Me has been deprecated. Please read the blog post for more info.

IDE Provisioner

A plugin compatibility check has been added to IDE Provisioner and is enabled by default. With this upgrade, all newly uploaded plugin versions are automatically verified for compatibility with the IDE, preventing the provisioning and installation of unsupported or mismatched plugins.

If end users experience issues with existing plugins (uploaded before the upgrade), you can run a manual compatibility check to identify and exclude the plugins causing the problems.

To run the compatibility check, execute one of the following API requests:

• POST /api/plugins/compatible-ides/recalculate-all

  Use this endpoint if you have fewer than 10,000 plugins (including all versions). Calling this endpoint triggers plugin compatibility calculation for all plugins uploaded before the upgrade.

• POST /api/plugins/compatible-ides/recalculate

  Use this endpoint if you have more than 10,000 plugins (including all versions). It allows you to trigger plugin compatibility calculation for specific plugins, helping to avoid server overload when determining compatible IDEs. Provide a list of plugin IDs in the request body, as shown in the example below:

  curl -X POST --location "https://<your_organization>/api/plugins/compatible-ides/recalculate" \ -H "Content-Type: application/json" \ -H "Authorization: Bearer <token>" \ -d '{ "pluginIds": ["aws.toolkit", "com.jetbrains.kmm"] }'