ReleaseBytes
github GitHub Changelog ·

GitHub Copilot for JetBrains adds persistent memory, Ollama integration, and enterprise controls

aigaengineer
feature patch

GitHub Copilot for JetBrains receives an update bringing persistent memory across chat sessions, enhancing context retention for developers. The release also integrates Ollama as a BYOK provider, offering more flexibility for local model access and configuration. Enterprise administrators gain new server-based controls for managing Copilot plugin availability and settings across their organizations. Additionally, the update includes expanded Codex workflow support, easier CLI setup, and various user experience and reliability improvements.

  • →Enterprise Managed Settings for Administrators
  • →Persistent Memory Across Chat Sessions
  • →Ollama as a Bring-Your-Own-Key (BYOK) Provider
  • →Expanded Codex Workflows and CLI Setup
  • →User Experience and Reliability Improvements
Features (3) ›
  • Enterprise Managed Settings for Administrators

    Administrators now have more server-based controls to manage GitHub Copilot across their organizations. These settings cover plugin availability, MCP server access, permission bypass behavior, and OpenTelemetry configurations.

  • Persistent Memory Across Chat Sessions

    Copilot memory can now retain and recall useful information across agent chat sessions, helping maintain context without repeatedly providing project details or preferences. This feature is manageable via a toggle in the Copilot settings portal.

  • Ollama as a Bring-Your-Own-Key (BYOK) Provider

    Developers can now use Ollama as a BYOK provider within GitHub Copilot for JetBrains, supporting provider configuration and model selection directly within the IDE experience for more tailored model integration.

Enhancements (2) ›
  • Expanded Codex Workflows and CLI Setup

    Codex sessions are now visible in agent debug logs and support updated permission modes and customizations through instructions and skills. Additionally, Copilot CLI can be automatically installed from integrated terminals on macOS, Linux, and Windows, simplifying setup.

  • User Experience and Reliability Improvements

    This update enhances account management flows, improves model and settings views, restores chat references for files and folders, and refines the placement of customization and debug log buttons. It also boosts reliability for MCP execution, terminal output, cloud agents, and fixes ANSI escape rendering.

Read the original announcement →

https://github.blog/changelog/2026-08-11-copilot-memory-and-ollama-in-github-copilot-for-jetbrains

Related releases