Skip to main content

First Run

Learn the basics of using Docket for your first AI conversation.

Understanding the Interface​

Docket Chat Interface

When you open Docket, you'll see:

  • Sidebar (left) — Search, New Chat, Models, Files, Offline/Lock toggles, and chat history
  • Chat Area (center) — Your conversations with AI models
  • Model Selector (bottom) — Choose your model and preset
  • Chat Settings (top right) — Adjust inference parameters

Starting Your First Chat​

1. Select a Model​

Click on the Models section in the sidebar to see available models:

ModelBest For
Llama 3.2 (1B)Fast responses, simple tasks
Gemma 3NGeneral purpose conversations
Qwen 2.5 UncensoredUnrestricted responses
Qwen 2.5 VL VisionImage understanding
Qwen 2.5 CoderProgramming assistance
Gemma 2Balanced performance
Qwen 2.5 14BComplex reasoning (requires more RAM)

2. Load the Model​

  1. Click on your chosen model
  2. Click "Load Model"
  3. Wait for the model to load (load times vary depending on computer specs)
  4. The status indicator will show "Ready" when complete

3. Send a Message​

  1. Type your message in the input field at the bottom
  2. Press Enter or click the send button
  3. Watch as the AI generates a response in real-time — completely offline and running locally on your machine

Tips for Better Results​

  • Be specific — Clear, detailed prompts get better responses
  • Use system prompts — Set context in Chat Settings
  • Adjust temperature — Lower for focused responses, higher for creativity
  • Try different models — Each model has unique strengths

Managing Chats​

  • New Chat — Click the "+" button in the sidebar
  • Rename — Double-click a chat title to edit it
  • Delete — Right-click a chat and select "Delete"

Next Steps​