Chat

Raycast offers two ways to work with an AI model. Quick AI handles fast, one-off questions straight from Root Search, while AI Chat is a full multi-turn conversation interface with persistent history, Agents, memory, attachments, and tool use. Together they cover everything from a quick lookup to a longer working session, all without leaving Raycast.

Models

Configure the AI models used across Quick AI, AI Chat, and Custom AI Commands from Settings → AI → Models, or use the Manage Models command to jump straight there.

Default Models

Pick the default model for each surface:

• Quick AI: the model used when you ask a quick question from Root Search.
• AI Chat: the default for new chats. Set to Last Used Model to continue with whichever model you used most recently.
• Custom AI Commands: the default for any AI Command that doesn't specify its own model.

New When a model supports reasoning effort, each row also shows a dropdown for it, with the recommended option marked Default. Your choice carries over between models that support it, or resets to the new model's default. The same control appears in Custom AI Command settings whenever a specific model is used instead of Last Used Model.

Manage Models

Choose which models appear in the model pickers across Quick AI, AI Chat, and Custom AI Commands:

• Review model details: each model lists its Speed, Intelligence, and Context window so you can compare capabilities at a glance.
• Enable or disable models: toggle the checkbox next to a model to control whether it shows up in pickers. Disabled models are hidden everywhere AI models can be selected.

• Group by Provider: models are grouped by their provider (e.g. Raycast, Anthropic). Expand or collapse a provider to focus on a subset.
• Sort the list: use the sort menu in the top-right to order models by Brand, Alphabetically, Speed, Intelligence, or Context Window.

Collapsed Messages

New To keep conversations easy to scan, long messages you send are collapsed in both Quick AI and AI Chat. A collapsed message shows its first part with a Show more control to expand it, and a Collapse control to fold it back once expanded.

Only long messages collapse: a message over 10 lines (or a comparably long single paragraph) is collapsed, while shorter messages are always shown in full.

Ask User Question

New Sometimes the AI needs a bit more direction. When a request is ambiguous or the model isn't sure how to proceed, it can pause and show you a short multiple-choice question right in the conversation. Pick an option and it continues with your answer. This works in both Quick AI and AI Chat.

To reliably trigger it, ask for it in your prompt: either name the tool directly ("use the ask questions tool") or describe the flow ("do X, then ask me to choose Y").

Try it: paste either of these into Quick AI or AI Chat:

Set up a new project folder for me and ask me to choose from the available options

Use the ask questions tool and quiz me about who are the cofounders of Raycast

Because these requests leave out details (where to create the folder and what to call it, or how many questions and how hard), the AI will typically ask a clarifying question with a few options before doing anything.

New You can also answer by voice: while a question or confirmation prompt is active, press Ctrl M to dictate your response. Learn more in Dictate in AI Chat.

Turning It Off

This is on by default. To disable it, open Settings → AI → General and turn off Ask Questions ("Allow AI to ask a short follow-up question before continuing"). With it off, the AI will make its best assumption and proceed without pausing to ask.

Quick AI

Quick AI is the fastest way to ask a one-off question from Raycast, without leaving your flow. Start from Root Search, hit Tab, and get an answer in the same window.

New Quick AI now uses a chat-style interface with follow-ups, context attachments, and a one-keystroke handoff to AI Chat.

Open Quick AI

There are three ways to start:

• From Root Search: start typing your question, then press Tab. Raycast hands your text to Quick AI and submits it.
• Run the Quick AI command directly from Root Search to open with an empty prompt.
• Set Quick AI as a fallback command in Settings → Launcher → Fallback Commands to send any unmatched Root Search text straight to Quick AI when you press ↵.

Pressing Tab from Root Search always opens Quick AI, even with no text typed, so you can start fresh.

Ask a Question

Type your prompt and press ↵ to submit. Quick AI streams the answer back in the same window.

On the response, ↵ pastes the answer into the previously focused app by default. To swap the default to copy instead, open Settings → AI → Commands → Quick AI → Primary Action.

When a response includes code, it appears in a code block with the detected language and a copy button. Long lines scroll sideways by default. Click Enable line wrap in the top right of the block to wrap them to the response width, and click again to switch back.

Follow-ups

Keep typing after a response to ask a follow-up. Quick AI is a full conversation, not a single-shot Q&A.

• Regenerate the last response with the same model: ⌘ / Ctrl R
• Regenerate the last response with a different model: ⇧ ⌘ / Ctrl R
• Start a fresh conversation: ⌘ / Ctrl N
• Navigate between recent Quick AI chats: ⌘ / Ctrl [ and ⌘ / Ctrl ]

Switch Models

Open the Action Panel (⌘ / Ctrl K) and pick Change Model to switch the active model mid-conversation, or Regenerate with Model… to try the last answer on a different model.

You can also set a default in Settings → AI → Commands → Quick AI → Default Model.

Continue in AI Chat

When a question grows into something bigger, press ⌘ / Ctrl J to move the conversation into AI Chat. You keep the full history and get the chat workspace, longer context, and tools.

Auto-new Chat

Quick AI starts a fresh chat after a period of inactivity, so you don't accidentally tack new questions onto an older conversation. Adjust the timeout in Settings → AI → General → Start New Chat (options: 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hour, always, never). This same inactivity timer also determines whether a Send to AI command reuses your active chat or starts a new one when AI Chat is closed.

Settings

Tune Quick AI from Settings → AI → Commands → Quick AI:

• Quick AI Default Model: default model used when opening Quick AI.
• Primary Action: paste to active app or copy the response to clipboard on ↵.
• Tab Shortcut: hide the Tab hint in Root Search (the shortcut still works).

AI Chat

AI Chat is Raycast's full chat workspace. Start a multi-turn conversation, attach files and context, give the model tools, and pick up where you left off across sessions.

New AI Chat is rebuilt in Raycast v2 with persistent history, Memory, rich attachments, and tool use, a full chat experience that replaces the single-shot AI flows in classic Raycast.

There are a few ways to start:

• Use the AI Chat command from Root Search to open the chat workspace.
• Press ⌘ / Ctrl J from a Quick AI conversation to continue it in AI Chat with the full history.
• Use a Send to AI command to send content into your active AI Chat from anywhere: Send Focused Window to AI, Send Selected Text to AI, Send Screen Area to AI, or Send Screen to AI. Learn more on the Screen Awareness page.

AI Chat lives in its own window, so you can keep it alongside whatever you're working on.

Start a Conversation

Type your prompt in the composer and press ↵ to send. ⇧ ↵ adds a new line. AI Chat streams the response back and remembers the whole thread, so each follow-up can build on earlier turns.

• Edit any previous message to re-run the conversation from that point.
• Regenerate the last response with ⌘ / Ctrl R.
• Start a fresh chat with ⌘ / Ctrl N.
• With an empty composer, ↑ / ↓ step through your recent messages.

When a response includes code, it appears in a code block with the detected language and a copy button. Long lines scroll sideways by default. Click Enable line wrap in the top right of the block to wrap them to the response width, and click again to switch back.

Steer or Queue Follow-ups

New While a response is streaming, you can Queue a follow-up to run when it finishes, or Steer to inject your message into the current run without restarting. The send button shows Steer Message or Queue Message for the active mode. Default is Queue. Change it under Follow-up Behavior in Settings.

• Do the opposite once: ⌘ / Ctrl ↵ applies the other behavior for one message (⌥ ↵ if that's your primary Send Message shortcut).
• Promote a queued message: click Steer on a queued message to inject it into the current run. With no active run, it simply sends.

Canceling is unchanged (empty input still cancels a run). Pending tool confirmations still pause the run; steered messages apply once it resumes.

Switch Models

Open the model picker from the composer. Pick a fast model for chat, a reasoning model for harder problems, or an image-capable model when you want to generate visuals.

Tune model behavior from the settings sidebar:

• Creativity: None, Low, Medium, High, or Maximum.
• Reasoning Effort: for models that support extended thinking.
• Web Search: let the model pull live results during the answer.

Memory

Turn on Memory (in Settings → AI → Personalization → Memory) and AI Chat will remember durable facts across conversations: your name, the projects you work on, the format you prefer answers in. The model writes to and reads from memory automatically as you chat.

Long Conversations

New Every model has a fixed context window, and a long working session will eventually fill it. Rather than cutting the conversation off, Raycast summarizes the earlier part of it in the background and carries on from the summary. You'll see a Summarizing chat… step in the transcript that settles into Summarized chat when it's done, and you keep going in the same chat.

Oversized tool results and attachments work the same way: Raycast saves the full content to a file and gives the model the start of it plus the path, so something too big for the context window can still be worked with. There's nothing to configure.

Attach Context

Add context to your next message by clicking the attach button in the composer or typing @. AI Chat can pull in:

• Files: Documents, code, or images from your disk.
• Notes: Any of your Raycast Notes.
• Clipboard History: A recent item you've copied.
• Browser Tabs: Open tabs from supported browsers.
• Web Search: Live web results gathered while answering.
• New Focused Window: A full Screen Awareness capture of the app in front of you, including its content, your selection, and a screenshot.
• Selected Text: Whatever you've highlighted in another app.
• Window…, Selected Area, or Entire Screen: A screenshot captured from your desktop.
• Calendar Events: Events from today or coming up.

Attachments are scoped to the message you send them with, but the model can keep referring back to them later in the chat.

Send to AI commands push content into your active chat instead of always starting a new one. The commands are Send Focused Window to AI, Send Selected Text to AI, Send Screen Area to AI, and Send Screen to AI.

What happens depends on AI Chat's state:

• AI Chat open: The content goes straight to the active chat.
• AI Chat closed: Your Start new chat after inactivity setting (see Auto-new Chat) decides: within the window the active chat is reused; once it's elapsed, a new chat begins.
• Mid-response: If you close AI Chat while the active chat is still streaming, that chat is always restored on next open, regardless of the timer.

Edit a Message

New Change a message you already sent and re-run the conversation from that point with the new text. Hover a previous message to reveal its actions, click the edit (pencil) icon, update the text in the composer (it shows Editing Message while you do), and resend.

Editing replaces the original response and any messages after it, so resending counts as a new AI request, just like regenerating a response.

Tool Use & Extensions

AI Chat can call tools: run a web search, generate an image, execute a terminal command, invoke an AI Extension, or hand work off to an MCP-compatible server. Toggle which tools are available per chat from the Tools section of the settings sidebar. With Tool Confirmation on, AI Chat asks before each tool runs.

New Generated images can be dragged and dropped straight from the chat into other apps.

Agents

Switch the chat to a saved Agent with the Change Agent action in the Action Panel. Agents set instructions, model, and AI Extensions for consistent results until you switch agennts or choose Ask Anything. See the Agents page page on creating your own.

Skills

Author reusable knowledge for the model with plain SKILL.md files. Raycast discovers them automatically and loads them when they're relevant to your chat. See the Skills page for the full format and where to put them.

Folders

New The history sidebar organizes your chats into sections: Pinned at the top, then any Folders you've made, then Recent (the default section for chats you haven't organized), and Archived at the bottom. Folders let you group related chats — a project, a topic, an ongoing investigation — so they don't get lost in Recent.

Create a folder in a few ways:

• Open the Action Panel (⌘ / Ctrl K) on any chat and choose Create Folder… (⌘ / Alt ⌃ / ⇧ N).
• Choose Move to Folder… (⌘ / Alt ⌃ / ⇧ M) from the Action Panel and pick New Folder… from the submenu.
• Right-click any chat in the sidebar and choose Move to Folder → New Folder… from the context menu.

The New Folder dialog lets you name the folder and pick an icon for it.

To rename or delete a folder, right-click the folder's header in the sidebar and choose Rename Folder or Delete Folder. Deleting a folder doesn't delete its chats — they move back to Recent.

Hover a folder's header to reveal a quick + button that starts a new chat directly inside that folder.

Manage Chats

The history sidebar shows every chat, with search at the top. From the Action Panel (⌘ / Ctrl K) on any chat — or by right-clicking the chat in the sidebar, which offers the same actions plus Copy Chat and Branch Chat — you can:

• Pin important conversations to the top.
• Archive chats you're done with. Old chats archive automatically after the window you pick in Settings (or never).
• Delete for good.
• New Rename a chat to give it a title you'll recognize, instead of the auto-generated one.
• New Move: drag a chat onto the Pinned section, a Folder, or the Archived section to reorganize it instantly, without opening a menu — or use Move to Folder… from the Action Panel or right-click menu. Dropping a pinned or foldered chat back onto Recent removes it from that pin or folder.

New You can also organize chats directly in the sidebar: drag and drop chats between sections like Recent, Pinned, Archive, and individual folders, or right-click anywhere in the sidebar to quickly create a new folder.

Always on Top

Toggle Always on Top from the Action Panel to keep AI Chat hovering over your other apps, handy while you're chatting about something on screen. You can also set this up in Settings → AI → Commands → AI Chat → Always on Top.

Background Notifications

New Raycast can notify you when an AI Chat finishes responding while it's running in the background, so you don't have to keep watching the window.

You'll get a notification when AI streaming finishes and the chat isn't in the foreground, meaning either the AI Chat is open but you're viewing a different chat, or the AI Chat window is blurred or minimized. The notification tells you why the chat finished (done, needs confirmation, failed, and so on), and clicking it takes you straight back to the chat it belongs to.

Enabling Notifications

Background notifications are off by default. The first time a chat finishes in the background, Raycast doesn't interrupt you. When you return to that chat, it shows a one-time alert asking whether you'd like to enable background notifications:

• Choose to enable, and Raycast turns on the toggle in Settings. On macOS, Raycast also shows the system permission prompt.
• Choose not to (or dismiss the alert with Esc), and the feature stays off and the alert won't appear again.

This onboarding alert is shown once per device.

Turning It Off

You can toggle AI Chat background notifications at any time in Settings. Turning it off silences only these notifications and leaves your other Raycast notifications intact.

Setting Up Notifications

To make sure background notifications reach you the way you want, it helps to check a few macOS system settings.

Permissions

Raycast needs permission to send notifications. The first time a chat finishes in the background, Raycast shows an onboarding alert when you return to it, and choosing to enable it triggers the macOS permission prompt. If you dismissed or denied it, or notifications still don't appear, open System Settings → Notifications → Raycast Beta and turn on Allow Notifications.

Focus Filters

If you use Apple's Focus feature, it's easy to accidentally filter Raycast out, so its notifications get held back while a Focus is on. If you have background notifications enabled but aren't seeing them, check whether your active Focus allows notifications from Raycast Beta in System Settings → Focus, and add it to the allowed apps if needed.

Show Notifications Permanently

By default, macOS shows notifications as Banners, which disappear on their own after a few seconds. If you don't want to miss that a chat has finished, set the alert style to persistent so notifications stay on screen until you dismiss them. You can change this in System Settings → Notifications → Raycast Beta.

If Notifications Don't Appear

Even with the toggle on, Raycast may not have system permission to send notifications. The toggle is a synced preference, but system permission is granted per device, so on another cloud-synced device the feature can be on without permission ever having been requested there. When permission is missing, Settings shows a hint explaining what's wrong, with a button to fix it:

• Permission not requested yet: the button shows the system permission prompt directly.
• Permission denied (the prompt was declined, or notifications were turned off manually in System Settings): the button opens your system notification settings, and Raycast rechecks the permission when you switch back.
• Notifications restricted by the system (for example on a managed device): the button also opens your system notification settings.

If permission is granted but notifications still aren't showing up, an active Focus may be holding them back. See Setting Up Notifications above.

Continue from Quick AI

When a Quick AI question grows into something bigger, press ⌘ / Ctrl J to move it into AI Chat. The full history, model, and any attachments come with it.

Settings

Tune AI Chat from Settings → AI:

• Start New Chat: start a new chat after the specified timeout when opening Quick AI, and determines whether a Send to AI command reuses the active chat or starts a new one when AI Chat is closed.
• Send Message: keyboard shortcut used to send messages in AI Chat (↵ or ⌘ / Ctrl ↵).
• Conversation History: show conversations from AI Chat and Quick AI combined or keep them separate.
• Auto-Archive Chats: automatically archive chats after the specified period of inactivity.
• Always on Top (per-command, under AI Chat): keep the AI Chat window above other windows.
• New Zoom: adjust the zoom level of the AI Chat window.
• New Follow-up Behavior (Advanced): queue follow-ups while the agent runs, or steer the current run. The alternate send shortcut does the opposite for one message.

Troubleshooting

AI Chat and Quick AI let you talk to models from anywhere in Raycast. Most reports come down to the trigger key, model selection, sync, or hitting a usage or context limit. Here are the common ones.