Chat Interface

Test your AI configurations with the built-in Chat Conversations. Experience real-time conversations and verify your training is working correctly.

Starting a Chat

There are several ways to start a chat session:

  1. From the configuration page, click Start Chat
  2. From the dashboard, click the chat icon on any configuration card
  3. From an existing chat, click New Chat in the sidebar

Chat Interface

The chat interface consists of:

  • Sidebar: Lists all your chat sessions for this configuration
  • Messages Area: Displays the conversation history
  • Input Area: Where you type your messages

Keyboard Shortcuts

Shortcut Action
Enter Send message — on a phone or tablet it is a new line instead, and the send button sends
Shift + Enter New line
⌘ / Ctrl + Enter Send message, on every device (the send button does the same)

File Attachments

You can attach files to your messages for the AI to analyze:

  • Images: JPEG, PNG, GIF, WebP (for vision-capable models)
  • Documents: PDF, Word, Excel, CSV
  • Code Files: Python, JavaScript, Ruby, HTML, CSS, etc.

To attach a file:

  1. Click the paperclip icon in the input area
  2. Select one or more files (max 20MB each)
  3. Preview appears above the input
  4. Type your message and send

Vision Support

Image analysis requires a vision-capable model like GPT-4o, Claude 3, or Gemini. Check your model's documentation for image support.

Message Features

Markdown Rendering

AI responses support rich Markdown formatting:

  • Headers, bold, and italic text
  • Code blocks with syntax highlighting
  • Bulleted and numbered lists
  • Tables
  • Links

Code Blocks

Code in responses is displayed with syntax highlighting. Hover over a code block to reveal the copy button.

Token Usage

Each AI response displays the number of tokens used. This helps you understand API costs and stay within model limits.

Chat History

All chat sessions are automatically saved. The sidebar shows:

  • Chat title (auto-generated from first message)
  • Message count
  • Last activity time

Click any chat in the sidebar to load it. You can have multiple ongoing conversations with the same configuration.

Deleting Chats

To delete a chat, click the trash icon next to it in the sidebar. This action is permanent and cannot be undone.

Conversation Context

The AI maintains context from earlier messages in the conversation. By default, the last 50 messages are included for context.

Each conversation is independent—the AI doesn't remember other chat sessions or conversations with other users.

Troubleshooting

Response Errors

If you see an error message:

  • Check that your API key is valid and has sufficient credits
  • Verify the configuration is set to "Active"
  • Try reducing the message length for very long inputs
  • Check the error message for specific details

Slow Responses

Response time depends on:

  • The AI model (larger models are slower)
  • Response length (controlled by max_tokens)
  • Provider API load

The "Thinking..." indicator shows that the request is being processed. Most responses complete within 5-30 seconds.