Assistant CoreAssistant Core
Web App

Chat Interface

Send messages, attach files, and interact with AI responses.

Sending Messages

  1. Type your message in the input bar at the bottom
  2. Press Enter or click the ↑ Send button
  3. The AI responds in real time — text streams character by character

Press Shift + Enter to add a new line without sending.

Attaching Files

You can send files along with your message for the AI to analyze:

Supported file types

TypeExamples
ImagesJPG, PNG, GIF, WebP
DocumentsPDF
VideoMP4, WebM
AudioMP3, WAV, WebM

How to attach

  • Click the 📎 attachment button in the input bar
  • Drag and drop files anywhere in the chat area
  • Paste an image from your clipboard (Ctrl/Cmd + V)

Attached files appear as thumbnails above the input field. Click on a file to remove it before sending.

File upload availability depends on your assistant's configuration. If the attachment button is not visible, file uploads are not enabled for this assistant.

Knowledge Base Toggle

If the assistant has a Knowledge Base configured, you'll see the 📖 Chat Doc button in the input bar:

  • Enabled (green) — The AI searches your uploaded documents when answering
  • Disabled (gray) — The AI uses only its general knowledge

Toggle it on when you want answers from your specific documents.

Thinking Mode

Some Chat models support thinking levels to spend more effort on harder tasks. When available, you'll see a 💭 Think button in the input bar:

LevelBehavior
OffStandard response, no thinking shown
LowBrief reasoning before answering
MediumModerate step-by-step reasoning
HighDeeper reasoning for math, logic, coding, or long analysis

Thinking mode is only available for supported Chat models. If the button doesn't appear, the current model does not support thinking. See Chat Model Selection.

Dictation

Click the 🎤 microphone button in the input bar to dictate your message:

  1. Click the mic — your browser will ask for microphone permission
  2. Speak your message
  3. Click the mic again to stop
  4. The transcribed text appears in the input field — edit if needed, then send

This is different from Voice Chat — dictation just converts speech to text for the input field.

AI Responses

Streaming

Responses appear in real time, character by character. While the AI is responding:

  • A streaming indicator appears
  • The ↑ Send button becomes a ⏹ Stop button — click to stop generation

Markdown & Code Blocks

The AI formats responses with:

  • Bold, italic, strikethrough
  • Bullet and numbered lists
  • Tables
  • Code blocks with syntax highlighting and a Copy button
  • Blockquotes and links

You can disable markdown rendering in Settings → General → Display Markdown.

Follow-up Suggestions

After each response, the AI may suggest follow-up questions. Click any suggestion to send it immediately.

Suggestions cost additional credits. You can disable them in Settings → General → Show Suggestions.

Message Actions

Hover over any AI response to see action buttons:

ActionIconDescription
Copy📋Copy the full response to clipboard
Like👍Rate the response positively — choose a reason (Well written, Accurate, Helpful, etc.)
Dislike👎Rate the response negatively — choose a reason (Incorrect, Too wordy, etc.)
Read Aloud🔊Listen to the response using text-to-speech

Feedback helps improve the AI's responses over time.

Scroll to Bottom

When you scroll up in a long conversation, a button appears near the input bar. Click it to jump back to the latest message.

On this page