> For the complete documentation index, see [llms.txt](https://morphic.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://morphic.com/docs/getting-started/copilot/chat-modes.md). # Chat modes Chat modes on Morphic let you decide how much control you have over each generation in [Copilot](/docs/getting-started/copilot.md). The mode is set per chat, so one chat can stay on automatic for quick exploration while another waits for your approval on expensive work. | Mode | What happens | | ---------------------- | ----------------------------------------------------------------- | | Automatically generate | Generates immediately, without stopping to confirm. The default. | | Ask before generating | Shows you the settings and the cost, and waits for your approval. | | Plan only | Researches and writes a plan, but never generates media. | ## Switch modes 1. Open [Copilot](https://studio.morphic.com/chat) 2. Click the mode button at the bottom right of the prompt bar. It shows the current mode as 'Auto', 'Ask', or 'Plan' 3. Under 'Mode', pick the one you want. A checkmark marks the active mode The mode stays set for that chat until you change it. Switch while Copilot is working and the change applies from your next message. ## Automatically generate The default. Copilot picks the settings, runs the generation, and returns the result without confirming anything. Each finished step reports how many credits it consumed. Best for exploring and iterating quickly, where a confirmation step would only slow you down. Since nothing pauses, a generation you did not intend still costs credits. The stop button halts a run but does not refund one already under way. When that matters, use 'Ask before generating'. ## Ask before generating Copilot prepares the generation, then waits for you. Nothing is charged until you approve. Best when credits matter: video, long durations, or a shared team balance. Here's a quick tutorial on how to use Ask before generating on Morphic: {% embed url="" %} 1. Set the mode to 'Ask before generating' and send your prompt 2. Copilot pauses and marks the chat 'Response required' in the sidebar, so you can find it later 3. Review the prompt and the settings, listed on one line: the model, aspect ratio, resolution, and whether web search is on 4. Click 'Generate' to run it, or 'Cancel' to drop it. The button itself carries the cost, reading 'Generate with' and the number of credits, and a bar above the prompt bar repeats it Cancel, and the step reads 'Generation cancelled' with nothing charged. Copilot does not retry on its own; it tells you what was skipped and offers it back with one click. {% hint style="warning" %} Cancelling is not the same as stopping. Cancelling at the approval step is free, because the generation has not started. Stopping one that is already running is not: the step reads 'User cancelled the operation', you get no output, and the credits are still spent. The approval step is the only point where backing out costs nothing. {% endhint %} Also worth knowing: * Approving costs exactly what it says: the figure on the button is what gets charged, and the finished step reports the same number back. * Free steps do not ask: operations that cost nothing, like cropping an image, trimming a clip, joining clips, or extracting audio, run straight through. * You can approve some and cancel others: each generation gets its own request and its own cost. The bar above the prompt bar shows the running total for everything still waiting, and its buttons act on all of it. {% hint style="info" %} You can ask what something costs in any mode. Try 'how many credits would a 10 second video cost with this model?' {% endhint %} ## Plan only Copilot researches your request and returns a plan without generating anything, so nothing is spent. Best for scoping a project, comparing approaches, or producing a cost estimate for a client. It can still read your attachments, search the web, look up your workflows, check what a model supports, and work out what the job would cost. It cannot create media, or create and edit workflows. 1. Set the mode to 'Plan only' 2. Describe the work, for example 'plan a 15 second product video for a coffee brand' 3. Copilot returns a plan. For a video that means a shot-by-shot breakdown, the approaches worth considering, a credit estimate by step, and the decisions it needs from you 4. Refine it in the same chat, then switch to 'Automatically generate' or 'Ask before generating' to run it ## Good to know * New chats start in the mode set as your default, which you pick in ['Preferences'](https://studio.morphic.com/settings/preferences). Changing one chat's mode does not change that default. * The mode belongs to the chat, not your account, so chats can be in different modes at once. * A new chat starts in your default mode, not the mode of the chat you were last in. * The approval step is the only free way out of a generation. After that, credits are spent whether or not you keep the result. * Chat modes apply to Copilot. Generating on the canvas is not affected. That's how chat modes work on Morphic. Pick the level of control the work needs, and change it whenever the work changes. *** If you have any trouble or have any questions, feel free to reach out. We're happy to help. Write to us at [**support@morphic.com**](mailto:support@morphic.com)**.** --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://morphic.com/docs/getting-started/copilot/chat-modes.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.