Skip to main content
ZeroTwo’s chat interface is designed to feel natural — type a message, press Enter, and get an AI response in real time. This guide walks you through everything from opening a new conversation to following up with questions, attaching files, and sharing your results.
Your first conversation uses the default model set in Settings → Preferences. You can change the default at any time, or switch models for any individual conversation without affecting your default.

​
Starting a New Chat

There are several ways to open a fresh conversation:
  • Click the pencil / compose icon at the top of the left sidebar
  • Press Ctrl/Cmd+Shift+O from anywhere in the app
  • Click New Chat on the welcome screen if no conversations are open yet
A blank prompt bar will appear at the bottom of the screen, ready for your first message.

​
Sending Your First Message

1

Click the prompt bar and type your message

The prompt bar is the text input at the bottom of the chat area. Click anywhere inside it to focus, then type your question or request in plain language. ZeroTwo understands natural language — no special syntax is needed for basic questions.Use Shift+Enter to add a new line without sending (useful for multi-paragraph prompts).
2

Press Enter to send

Press Enter to send your message, or click the send button (arrow icon) on the right side of the prompt bar. The currently selected model name is visible in the prompt bar — this is the model that will respond.
3

Watch the response stream in real time

The AI response streams word by word as it is generated. Code blocks receive automatic syntax highlighting, and markdown formatting (headers, bold, bullet lists, tables) renders immediately.
4

Ask a follow-up question

ZeroTwo retains the full context of your conversation. Just type your follow-up directly — there is no need to repeat context from earlier in the chat.
Can you give me a Python example of that?

Simplify the last explanation for someone non-technical.

What are the trade-offs of that approach?

​
What You Can Ask

ZeroTwo handles a wide range of tasks. Here are some examples to get you started:
TaskExample prompt
Writing”Write a professional email declining a meeting request.”
Coding”Explain what this Python function does and suggest improvements.”
Research”What are the key differences between RAG and fine-tuning for LLMs?”
Analysis”Summarize the main arguments in this article and identify any logical gaps.”
Brainstorming”Give me 10 creative names for a SaaS product that helps teams manage async video updates.”
Translation”Translate the following paragraph into formal Japanese.”
Math & Logic”Walk me through how to solve this linear algebra problem step by step.”
Data”I have a CSV with sales data. What trends should I look for?”

​
Message Actions

Once an AI response appears, hover over it to reveal action buttons:
ActionWhat It Does
CopyCopies the full response text to your clipboard
ShareCreates a public read-only share link for the conversation
RegenerateRe-runs the same prompt to get a different response — you can optionally switch to a different model before regenerating
Thumbs UpMarks the response as helpful
Thumbs DownMarks the response as unhelpful
For your own messages, hover to reveal an Edit button. Editing a message regenerates the AI response from that turn forward; all subsequent messages in the thread are replaced.

​
Attaching a File

You can attach files to any message to ask ZeroTwo questions about them:
  1. Click the paperclip icon in the prompt bar, or drag and drop a file directly onto the chat area
  2. A file preview thumbnail appears above the text input
  3. Type your question about the file (or leave blank — ZeroTwo will infer what to do)
  4. Press Enter to send
Example prompts after attaching a file: 
Summarize the key findings from this PDF.

What does this Python function do? Can you refactor it?

Transcribe this audio and list the action items mentioned.
ZeroTwo remembers your conversation — ask follow-up questions about an uploaded file in subsequent turns without re-attaching it. The file context persists throughout the entire chat session.
Supported file types include PDFs, Word documents, images, audio, video, code files, CSV, JSON, and more. For full details see Upload Your First File.

​
Switching Models

Click the model name in the prompt bar at any time to open the model picker. Selecting a new model applies it to all subsequent messages — previous responses in the thread remain unchanged. You can also switch models specifically for a Regenerate action: after clicking Regenerate, a model selector appears so you can compare how different models respond to the same prompt. Learn about all available models →

​
Using @ Mentions

Type @ in the prompt bar to invoke a connected app or skill. A dropdown appears listing all your active connectors and skills. Examples:
  • @google-drive — reference files from your Google Drive in the current message
  • @github — access your GitHub repositories, issues, and pull requests
  • @notion — pull content from your Notion workspace
To connect apps, go to Settings → Connectors. See Settings Overview for details.

​
Sharing a Conversation

To share a chat with someone else:
  1. Hover over any AI response and click Share, or use the share button in the chat header
  2. ZeroTwo generates a public read-only link
  3. Copy the link and send it — the recipient does not need a ZeroTwo account to view a shared chat
You can manage and revoke all your shared links from Settings → Data Controls.

​
Next Steps

Choose a Model

Browse 95+ models and learn the difference between standard, premium, and reasoning categories.

Choose a Tool

Enable Web Search, Deep Research, Canvas Mode, Agent Mode, or Voice for your next message.

Upload Your First File

Learn which file types are supported and how ZeroTwo processes each one.

Home Page Tour

Get a full tour of the ZeroTwo interface — sidebar, prompt bar, topbar, and panels.