Continue v1.0.65 fixes JetBrains IDE freezes, adds DeepSeek Reasoner support
Continue released v1.0.65 for JetBrains IDEs and v1.3.36 for VS Code, focusing on stability fixes. The update addresses IDE freezing issues in JetBrains, adds compatibility with DeepSeek's reasoning model, and fixes multiple memory leaks and crash conditions.
Continue v1.0.65 Fixes JetBrains IDE Freezes, Adds DeepSeek Reasoner Support
Continue, the open-source AI coding assistant, released v1.0.65 for JetBrains IDEs and v1.3.36 for VS Code, prioritizing stability over new features. The update targets multiple crash and freeze conditions that affected JetBrains users.
Key Fixes
JetBrains Stability: The release hardens remote config sync to prevent IDE freezes and chunks large JCEF (Chromium Embedded Framework) messages to prevent sidebar freezes. These were critical stability issues affecting the plugin's usability in JetBrains products.
DeepSeek Reasoner Compatibility: Continue now maps max_tokens to max_completion_tokens for DeepSeek's reasoning model, enabling proper integration with OpenAI-compatible API adapters. This fix allows users to leverage DeepSeek's chain-of-thought reasoning capabilities through the Continue extension.
Memory and Crash Prevention: The update resolves multiple technical issues:
- Memory leak in CoreMessenger's responseListeners
- DiffStreamHandler crashes from unsafe type casting and negative line number handling
- Disposed JCEF browser reference guards in sendToWebview
- SideEffectNotAllowedException in intention previews
- Message type pass-through synchronization between core and JetBrains implementations
Test Updates
Binary test expected files were updated to reflect architectural changes in the JetBrains implementation, ensuring test coverage remains current with the codebase.
Dependency Updates
The continue-sdk package bumped brace-expansion from 5.0.4 to 5.0.5, addressing a minor dependency security advisory.
Version Alignment
With JetBrains at v1.0.65 and VS Code at v1.3.36, the extensions continue on separate version tracks reflecting their distinct codebases and release cadences. Both share the same core Continue engine but maintain separate UI implementations for their respective IDEs.
What This Means
This patch release signals Continue's focus on reliability over feature velocity. JetBrains users experienced real stability problems—IDE freezes are showstoppers—making these fixes essential for production use. The DeepSeek Reasoner integration is timely, as reasoning models gain adoption among developers. These aren't flashy features but they address the gap between Continue's promise and actual user experience in enterprise IDEs like IntelliJ IDEA, PyCharm, and GoLand.
Related Articles
GitHub Models Retired, Cutting Off Free LLM API Access for GitHub Actions
GitHub Models, the unified LLM API and playground that let GitHub Actions workflows make free AI calls using existing GitHub credentials, has been fully retired. GitHub has not disclosed a reason, though rising inference costs from coding-agent usage are the likely driver.
Anthropic Makes Claude Code's Auto Mode Default for Pro, Max, and Team Users on August 14
Anthropic will make Claude Code's auto mode the default for Pro, Max, and Team accounts starting August 14, reducing step-by-step approval prompts. The company cites a study of 1,053 testers showing auto mode caught 89% of harmful actions versus 13.6% for manual review.
Anthropic Makes Auto Mode Default in Claude Code for Pro, Max, and Team Plans Starting August 14
Anthropic will make auto mode the default setting for new Claude Code sessions on Pro, Max, and Team plans starting August 14, 2026. The company cites a 1,053-person study showing auto mode blocked 89% of harmful actions compared to 13.6% for human reviewers, plus a third-party test claiming zero successful prompt injections out of 720 attempts.
Anthropic Makes Claude Code's Auto Mode Default, Cutting Dangerous Command Approvals by 75 Percentage Points
Anthropic will enable Auto Mode by default in Claude Code for Pro, Max, and Team plans starting August 14, 2026. Internal testing with 1,053 paid users found the classifier-based system caught 89% of dangerous commands compared to 13.6% caught by human reviewers, while producing 25% more pull requests.
Comments
Loading...