Beginners
Chapter 3: Choosing LLM for Cline
Learn to pick the best LLM in Cline by balancing speed, context, cost, and flexibility—matching models to your specific tasks
Latest
Learn
Chapter 2: LLM Benchmarks
Learn how to interpret LLM benchmarks—coding, domain, and tool-use—to match scores with your development needs.
Learn
Chapter 1: LLM Fundamentals
Learn how to choose the right LLM in Cline by understanding trade-offs in speed, cost, reasoning, and multimodality.
Updates
AI Coding University: Core Skills for Building with AI Coding Agents
We’re launching AI Coding University to help developers build the core skills for agentic coding.
Updates
Cline v3.28: Free Grok Extended + GPT-5 Optimized
xAI extends free Grok access while we ship GPT-5 optimizations that actually matter for real coding tasks.
Learn
Chapter 1: Prompt Fundamentals
Cline works best when your ask is specific. In this blog we review strategies to prompt Cline in an effective way.
Learn
Chapter 2: Cline Rules
Learn how System Prompt Rules guide coding style, security, and workflow ensuring consistency, safety, and alignment in every task.
Learn
Chapter 3: System Prompt Fundamentals
Learn how System Prompt bridges intent and action with three pillars: Tools, System Info, and User Preferences.
Learn
Chapter 4: System Prompt Advanced
Learn how System Prompt Advanced handles dev tasks: explore code, make diff edits, verify with commands, and test in the browser.
Guides
Stop Adding Rules When You Need Workflows
Most developers clutter their context with rules when they should build workflows. Workflows inject instructions on-demand and consume tokens only when invoked, while clinerules append to every system prompt.
Models
Sonoma Alpha Sky & Dusk Models: Real-World Performance in Cline
Two mysterious models with 2M context windows appeared on major AI gateways. We tracked their performance across thousands of real coding tasks to see if the hype matches reality.
Models
Kimi K2-0905: Moonshot’s Latest Open-Source Model is Live in Cline
Kimi K2-0905 doubles the context window to 256k tokens and delivers some of the highest reliability tool calling we've seen in an open-source model. Built for coding agents that actually work.