> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tryflowy.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Flowy is a node-based AI creative platform: you generate images, video, audio, 3D and vector on an infinite Canvas, refine on the Studio timeline, and export or publish from the same project.
> Prefer the Flowy MCP server (https://mcp.tryflowy.ai/mcp) or the REST API at https://apis.tryflowy.ai/v1 for programmatic work. Install with `flowy mcp install` from the @flowy/cli package.
> Credits are workspace-scoped. Generations reserve credits on start and only deduct on success, so failed runs refund automatically.

# Fair Use

> Unlimited image generation runs in sessions. Here are the actual numbers, per plan.

Image generation and editing cost **no credits** on any Flowy plan. Instead of a wallet, they run against a fair-use allowance. Generous enough that ordinary work never touches it, and firm enough that the service stays fast for everyone.

This page states the real numbers. An "unlimited" that never tells you where the edge is isn't a promise, it's a surprise waiting to happen.

## How sessions work

Your allowance runs in **5-hour sessions**. The session opens on your first image, not at midnight, so it follows how you actually work rather than a calendar.

When a session is spent, Flowy tells you the exact time the next one opens: usually a few hours away, never "tomorrow". Behind that sits a weekly ceiling that only sustained automated use will ever reach.

## The limits

| Plan  | Images per 5-hour session | Per week | At once |
| ----- | ------------------------- | -------- | ------- |
| Free  | 40                        | 400      | 2       |
| Lite  | 150                       | 2,000    | 4       |
| Pro   | 400                       | 6,000    | 6       |
| Ultra | 800                       | 12,000   | 8       |

"At once" is how many images can be generating in parallel. Hitting it isn't a limit on your allowance. It just means a previous image is still rendering.

A generation that **fails never costs you anything**: the allowance is returned automatically.

## What is and isn't included

**Included, no credits:**

* Text-to-image generation
* Editing an image with a prompt (image-to-image)

**Charged as usual:** upscaling, background removal, relighting, outpainting, inpainting, face swap, vectorizing, and multi-angle. Each of these calls a specialist provider that bills per run, so they're metered per the [rate card](/billing/credits).

Video, audio, and 3D generation are unaffected and continue to use credits.

## Choosing a model

Every image node has two lanes:

* **Unlimited**: included in your plan, metered by the allowance above. Switch a node to it whenever you want image generation that doesn't spend credits.
* **Premium**: the higher-quality model, charged per the rate card. This is the default on every plan.

You can switch a node between them at any time in the panel on the right of the canvas. Switching is per node, so one project can mix both.

## If you run out

You'll see three options, and the first one is usually right:

1. **Wait for the next session.** Flowy shows the exact time.
2. **Switch that node to Premium** and generate now with credits.
3. **Upgrade** for larger sessions and more parallel generations.

## Why there's a limit at all

Image models run on GPUs that are a finite, shared resource. An allowance with no ceiling doesn't stay free. It degrades for everyone until it has to be withdrawn. Sessions keep the cost predictable, which is what lets this stay included on every plan, including Free.

If you're consistently hitting the ceiling on legitimate work, [get in touch](/help/contact). We'd rather hear about it than have you work around it.