> For the complete documentation index, see [llms.txt](https://help.imagine.art/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.imagine.art/workflows/quick-mode.md).

# Quick Mode

Quick Mode is the fastest way to generate content in ImagineArt. Type a prompt, adjust a few settings, and hit generate - no nodes, no connections. Everything happens from a single prompt bar at the bottom of the screen.

If you've used any AI generation tool before, this will feel familiar. Describe what you want, optionally add reference media, and let the model do the rest.

***

### How to Switch to Quick Mode

Look at the bottom toolbar of your workspace. You'll see two mode icons on the right side:

<table><thead><tr><th width="222.625">Icon</th><th width="216.0625">Mode</th><th>What It Does</th></tr></thead><tbody><tr><td><strong>⚡ Lightning bolt</strong></td><td><strong>Quick Mode</strong></td><td>Single prompt bar. One input, one output. The fastest path to a generation.</td></tr><tr><td><strong>🔀 Nodes</strong></td><td><strong>Workflow Mode</strong></td><td>Full canvas with nodes, connections, and multi-step pipelines. For complex, multi-model workflows.</td></tr></tbody></table>

Click the **lightning bolt** icon to switch to Quick Mode. Click the **nodes** icon to switch back to the full Workflow canvas.

> **When to use which:** Quick Mode is for straightforward, single-step generations — "I want an image of..." or "Turn this photo into a video." Workflow Mode is for multi-step pipelines where you need to chain models, edit outputs, or build reusable automations.

***

### The Prompt Bar

The prompt bar sits at the bottom center of the screen. This is where all your inputs and settings live.

#### Writing a prompt

Click the text field ("Ask anything...") and describe what you want to generate. The prompt is sent directly to the selected model - what you write is what drives the output.

#### Adding references

Above the prompt field, you'll see three reference buttons:

<table><thead><tr><th width="220.76171875">Reference Type</th><th>What It Does</th></tr></thead><tbody><tr><td><strong>Reference Images</strong></td><td>Upload images to guide the generation — style references, character photos, product shots, or any visual input the model should work from.</td></tr><tr><td><strong>Reference Videos</strong></td><td>Upload video clips as reference material for video generation models.</td></tr><tr><td><strong>Reference Audio</strong></td><td>Upload audio files for models that support audio input (voiceovers, music, lip-sync).</td></tr></tbody></table>

Not every model uses every reference type. The available options depend on which model you've selected in the settings bar.

***

### Settings Bar

Below the prompt field, a compact settings bar lets you configure the generation without opening a separate panel. Everything is inline — click any setting to change it.

#### Reference

Opens the reference configuration panel. This is where you manage uploaded reference media and adjust how the model uses it (strength, style influence, etc.).

#### Model

Shows the currently selected AI model (e.g., **Seedance 2.0**). Click to switch between available models. Different models support different output types and have different strengths — some excel at photorealism, others at animation or stylized content.

#### Output settings

Displayed as a compact string like **16:9 / 1080p / 5**, this controls:

| Setting              | What It Controls                                                                                                          |
| -------------------- | ------------------------------------------------------------------------------------------------------------------------- |
| **Aspect ratio**     | The shape of the output — 16:9, 9:16, 1:1, 4:3, and other ratios depending on the model.                                  |
| **Resolution**       | The output quality — 720p, 1080p, or higher where supported.                                                              |
| **Duration / Steps** | For video models, this sets the clip length in seconds. For image models, this may control the number of inference steps. |

Click the settings string to open a picker and adjust each value.

#### Multi-prompts

A toggle that enables multi-prompt input. When turned on, you can enter multiple prompts that the model processes in sequence or combination — useful for storyboarding, scene transitions, or giving the model layered instructions.

#### Generation count

Shown as a counter (e.g., **4/20**) with **−** and **+** buttons. This controls how many outputs the model generates per run. Increase the count to get more variations from a single prompt; decrease it to save credits.

***

### Generation Types

The bottom toolbar also lets you switch between different generation types. Each icon selects what kind of content you're creating:

| Icon      | Type                 | What It Generates                                                         |
| --------- | -------------------- | ------------------------------------------------------------------------- |
| 🖼️ Image | **Image generation** | Generate still images from text prompts and reference media.              |
| 🎥 Video  | **Video generation** | Generate video clips from prompts, reference images, or reference videos. |
| 🎙️ Audio | **Audio generation** | Generate audio, voiceovers, or sound effects.                             |

The active type determines which models and reference options are available in the settings bar.

***

### Where Results Appear

Generated outputs appear on the canvas area above the prompt bar. Each generation creates a new result that you can:

* **Preview:** View the full-size output directly on the canvas.
* **Save:** Results are automatically saved to your project's **Content** tab in the left sidebar.
* **Use as reference:** Drag a generated result back into the prompt bar as a reference for the next generation.
* **Send to Workflow:** Take a result you like and build a full multi-step workflow around it in Workflow Mode.

***

### Quick Mode vs. Workflow Mode

<table><thead><tr><th width="186.14453125"></th><th>Quick Mode</th><th>Workflow Mode</th></tr></thead><tbody><tr><td><strong>Interface</strong></td><td>Single prompt bar inside canvas</td><td>Full node canvas</td></tr><tr><td><strong>Best for</strong></td><td>Fast, single-step generations</td><td>Multi-step pipelines with chained models</td></tr><tr><td><strong>Inputs</strong></td><td>One prompt + optional references</td><td>Multiple nodes, each with their own parameters</td></tr><tr><td><strong>Outputs</strong></td><td>Direct to canvas</td><td>Routed through connected nodes</td></tr><tr><td><strong>Learning curve</strong></td><td>Immediate — type and generate</td><td>Requires understanding nodes and connections</td></tr><tr><td><strong>App Builder</strong></td><td>Not available</td><td>Build and publish apps from workflows</td></tr></tbody></table>

> **They work together.** Quick Mode is great for exploring ideas fast. Once you find a result or approach you like, switch to Workflow Mode to build a repeatable pipeline around it and optionally package it as an app with App Builder.

***

### Tips

* **Start broad, then refine.** Use Quick Mode to explore different prompts and models quickly. Once you find a direction, you can increase the generation count for more variations or switch to Workflow Mode for fine-tuning.
* **Use references for consistency.** If you want outputs that match a specific style, character, or product, always upload a reference image. Prompts alone can drift — references anchor the generation.
* **Try different models.** Each model has a different personality. Seedance, Flux, and other available models will interpret the same prompt differently. Switching models is one click.
* **Watch your credits.** The generation count multiplies the credit cost per run. Generating 10 outputs costs 10× what generating 1 costs. Start with a low count while experimenting, then scale up when you're confident in the prompt.


---

# 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://help.imagine.art/workflows/quick-mode.md?ask=<question>&goal=<endgoal>
```

`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.