GitHub Copilot for JetBrains: new version adds agent memory, local models through Ollama and enterprise settings management
GitHub has released a new version of Copilot for JetBrains with agent memory across sessions, support for Ollama as a local BYOK provider, new enterprise managed settings and reliability fixes for MCP, the terminal and Codex workflows.
GitHub has released a new version of the GitHub Copilot plugin for JetBrains that adds persistent agent memory, access to local models and expanded enterprise controls. The new Copilot memory feature allows the agent to retain and reuse information across individual chat sessions, so users do not need to repeatedly enter the same project details or their preferences. The feature can be enabled or disabled using the Copilot Memory toggle in the Copilot settings portal.
The plugin now supports Ollama as a BYOK (bring-your-own-key) provider, allowing users to configure the provider and select a model directly in JetBrains and work with local models. Administrators also get server-side (enterprise managed) settings covering plugin availability, access to MCP servers, permission bypass behavior (permission bypass) and OpenTelemetry settings across the organization.
According to GitHub, the plugin can now automatically install Copilot CLI from integrated terminals on macOS, Linux and Windows, making it easier to launch agent workflows from the terminal directly in the IDE. Codex sessions are now visible in agent debug logs, and Codex workflows support updated permission modes and customization through instructions and skills.
The release also brings reliability fixes for MCP execution and approval, terminal output and automatic approval, customization, cloud agents and diff-based edits, including a fix for rendering ANSI escape sequences and more predictable terminal scrollbar behavior. User-facing product text now uses the label “Copilot” instead of “Copilot CLI”.
Why it matters
For individual developers, agent memory means less repetitive entry of context and a faster start when working with the agent; Ollama support makes it possible to use their own local models instead of cloud models. For companies, enterprise managed settings provide central control over who uses the plugin and MCP servers and how they use them, which is relevant when deploying Copilot in larger teams with permission and telemetry management requirements.
Two audiences, two different impacts
What this means
For individuals
Developers using JetBrains with GitHub Copilot no longer need to repeatedly enter the same project information or preferences, because the agent now remembers them across chat sessions, and they can also connect their own local models through Ollama.
For a business
Administrators get server-side (enterprise managed) settings that let them centrally control the availability of the GitHub Copilot plugin for JetBrains, access to MCP servers, permission bypass behavior and OpenTelemetry settings across the organization.
Risks and complianceCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.