Cline Releases Desktop App for Open-Weight Models, Supporting Parallel Sessions and Automation
Key point
Cline has released an early version of a desktop app designed to support open-weight models and manage parallel agents.
Details
Building on the success of its VS Code extension, Cline has released an early version of a dedicated Cline Desktop app for open-weight models. The app is designed to make it easier to delegate and manage AI agents for tasks beyond coding, including non-development work.
Model Choice Freedom and Compatibility
Users can access over 300 models via the Cline provider or use open-weight models with ClinePass. It also supports running local models and connecting more than 50 API keys.
- Separation of Plan and Act Stages: Different models can be applied to each stage.
- Task Continuation: Conversations and contexts from Claude Code or Codex can be imported to continue execution in Cline.
- Same Harness: The Mac app uses the same Cline harness as the CLI and VS Code extension, allowing tasks to be continued via session linking.
Parallel Sessions and Automation Features
Multiple agents can be run simultaneously to handle coding, research, and repetitive tasks in parallel. In particular, the Schedule feature allows agent tasks to be scheduled for execution like cron jobs.
- Automation Examples: Vulnerability scanning, critical bug detection, and maintaining documentation updates.
- Extensibility: Extend agent capabilities by installing tools, plugins, and MCP servers from the Marketplace.
- Input Methods: Supports Voice Input and Web Search to streamline information gathering and task efficiency.
Performance and Accessibility
Based on the Cline CLI, scores on the Terminal-Bench 2.0 evaluation (pass@1) were 60.67% for DeepSeek V4 Flash and 82.02% for Kimi K3. Cline Desktop is currently distributed as a beta version for Mac and Windows, and user feedback is being collected.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.