Skip to main content

Auto Mode

Auto Mode is an intelligent feature that automatically selects the best AI model for each task based on complexity, context, and requirements. With Auto Mode enabled, you no longer need to manually choose between different models—AI Cockpit analyzes your request and routes it to the optimal model automatically.

What is Auto Mode?​

Auto Mode uses intelligent routing to match your tasks with the most appropriate AI model. It evaluates factors like task complexity, context size, and specific requirements to determine which model will provide the best results while optimizing for both performance and cost.

tip

Auto Mode is perfect for users who want to focus on their work without worrying about which model to use for each task.

How It Works​

Auto Mode follows a four-step process to ensure optimal model selection:

  1. Analyzes your request and current context - Examines the task description, mentioned files, and workspace context
  2. Evaluates task complexity and requirements - Determines if the task is simple (quick questions, minor edits) or complex (architecture design, large refactoring)
  3. Selects the optimal model automatically - Chooses from available models based on the analysis
  4. Routes your request to the best available model - Sends your task to the selected model for processing

Benefits​

Smart Routing​

Auto Mode intelligently routes each request to the model best suited for the task, ensuring you always get optimal results.

Cost Optimization​

By automatically selecting cost-effective models for simpler tasks and reserving powerful models for complex work, Auto Mode helps optimize your API usage costs.

Best Performance​

You always get the best performance for your specific task without needing to understand the capabilities and trade-offs of each available model.

No Manual Selection​

Focus on your work instead of model selection. Auto Mode handles the technical decision-making for you.

Usage Examples​

Simple Tasks​

For straightforward tasks like quick questions, simple code changes, or basic refactoring, Auto Mode automatically selects faster, cost-effective models:

"Fix the typo in the button label"
"What does this function do?"
"Add a console.log statement here"

These tasks are routed to efficient models that provide quick responses at lower cost.

Complex Tasks​

For sophisticated work like architecture design, complex refactoring, or multi-file changes, Auto Mode automatically switches to more powerful models:

"Refactor this application to use a microservices architecture"
"Implement a new authentication system with JWT tokens"
"Analyze the entire codebase and suggest performance improvements"

These tasks are routed to advanced models with stronger reasoning capabilities.

Enabling Auto Mode​

Auto Mode can be enabled through the chat interface:

  1. Look for the Auto Mode toggle in the chat input area
  2. Click the toggle to enable Auto Mode
  3. The icon will change to indicate Auto Mode is active
info

When Auto Mode is enabled, the model selector will show "Auto" instead of a specific model name.

When to Use Auto Mode​

Auto Mode is ideal for:

  • General development work - Let AI Cockpit choose the right model for each task
  • Mixed complexity workflows - When you're working on both simple and complex tasks
  • Cost-conscious development - Automatically optimize costs without sacrificing quality
  • Learning and exploration - Focus on your work without worrying about model capabilities

When to Manually Select Models​

You might prefer manual model selection when:

  • Testing specific models - Comparing outputs from different models
  • Consistent model usage - Requiring the same model for a series of related tasks
  • Budget constraints - Strictly limiting usage to specific cost tiers
  • Model-specific features - Using capabilities unique to a particular model

Best Practices​

Provide Clear Context​

Auto Mode works best when you provide clear context about your task. Use Context Mentions to reference specific files or problems:

@/src/components/Button.tsx Refactor this component to use TypeScript

Break Down Complex Tasks​

Even with Auto Mode, breaking complex tasks into smaller steps helps ensure optimal results:

1. "Analyze the current authentication system"
2. "Design a new JWT-based authentication flow"
3. "Implement the new authentication system"

Review Model Selection​

While Auto Mode handles selection automatically, you can observe which models are being used by checking the API request information in the chat. This helps you understand how Auto Mode is routing your tasks.

Learn More​

For more information about model selection and configuration, see: