# Meet Sloosh: make images and video, then make them again

Sloosh is one workspace for images and video: prompt in Studio, run a Space someone already built, or build your own flow in Spacelab.

8 Oct 2026 · 3 min read · Product · By Sloosh

Web page: https://sloosh.ai/blog/meet-sloosh

Most AI image tools are good at one picture. Then you need the same look for the next product, and the one after that, and the prompt you used last week is gone. Sloosh is built for that second picture, and the hundredth.

## Start in Studio

Open Image or Video, describe what you want, and press Generate. You pick from 11 image models and 15 video models, including Nano Banana, GPT Image, FLUX, Seedream, Ideogram, Seedance, Kling, Veo and MiniMax. The chips under the prompt hold each model's own settings, like aspect ratio, resolution and how many to make.

The Generate button shows the credit cost before you press it, and nothing is spent until you do.

If you don't know what to write, open the Templates tab. Every card is a real result with the prompt and settings that made it. Press Recreate, swap in your own product photo, and generate.

**Starter prompt**

```text
Studio product shot of a matte black water bottle on wet slate, soft morning light from the left, shallow depth of field, a few water droplets on the cap, editorial and calm.
```

![A matte black water bottle with water droplets on wet slate, in soft morning light](https://cdn.sloo.sh/blog/introducing-sloosh/starter_prompt_water_bottle_7bfe2984bf.jpg)

_What that prompt makes with Nano Banana 2.1 at 16:9._

## Run a Space

A Space is a whole workflow behind a simple form. Someone built the steps once: the prompt, the model, the edits. You upload a product photo, press Generate, and get the same kind of result every time.

Reach for a Space when you need the same look across many products, or when someone already built the flow you need. For a one-off idea, Studio is quicker.

![The same tan leather backpack in four matching photos: a forest trail, a city street at dusk, a beach and a white studio](https://cdn.sloo.sh/blog/introducing-sloosh/backpack_four_scenes_fc59d39763.jpg)

_One product, four scenes, one consistent look._

> **Tip: Batch runs.** On Pro and Max, a Space runs in batch: give it a folder with one subfolder per product and it runs them all.

## Build your own in Spacelab

Spacelab is the canvas where Spaces are made. Each step is a node: a photo, a prompt, a model, an edit. Wire them together, run one step at a time to test it, then publish the flow as a Space with its own short link. Whoever runs it uses the form and never opens the canvas.

**Which one to use**

| Use | When |
| --- | --- |
| Studio | One prompt, one model, one result |
| Spaces | Run a workflow someone already built |
| Spacelab | Chain steps into a flow you will run again |

## Everything lands in Assets

Every image and video your workspace makes is kept in Assets. Save the ones you want to reuse as Elements and add them to your next prompt as references.

## Use it from Claude, Cursor or VS Code

Sloosh has an MCP server. Connect Claude, Cursor, VS Code or Codex to your workspace to run Spaces, generate, and build canvases from your editor.

## Try it

Your first workspace gets free starter credits. Open Studio, write one sentence, and see what comes back.

## FAQ

**Do I need to know prompting to use Sloosh?**

No. Start from a template or a Space: both fill in the prompt and settings for you.

**How do I know what a generation costs?**

The Generate button shows the credit cost before anything runs.

**Can my team use the same flows?**

Yes. Publish a Spacelab canvas as a Space, and you, your team or anyone you publish it for can run it from the form.