---
title: Image settings explained
display: '*Image* settings explained'
nav_title: Image settings
summary:
  - "Each image setting, its API parameter, its usual range, and which current models expose it."
  - "The panel shows only what the selected model accepts; its Parameters tab is the live source."
  - "Count, size, resolution and quality change the price; background, seed and magic prompt do not."
description: What each image setting does, its usual range, and which current models expose it, as of October 2026.
section: images
order: 20
type: reference
level: Build
roles: [all]
time: { read: 6 }
last_verified: 2026-11-11
models_checked: ["GPT Image 2.5 Sunburst", "GPT Image 2.5 Flare", "Nano Banana 2.1", "Seedream 5.0 Pro", "Ideogram 4.5", "Retro Diffusion Plus", "Topaz Image Upscale"]
related: [generate-images, image-models-guide]
video: tutorials/image-settings-explained.mp4
video_caption: Image settings explained
about: "The settings under the prompt decide how many images you get, their size, how much effort goes into each one and whether they have a background. This lesson changes them one at a time on GPT Image 2.5 Sunburst to make a transparent cutout of the tank hero from Generate images, then switches to Ideogram 4.5 to show how the settings change with the model. The models and prices in this lesson are as of November 2026."
chapters:
  - { at: 0, title: "The settings", image: "/media/images/images/image-settings-explained/ch-01-the-settings.webp" }
  - { at: 68.3, title: "Another model", image: "/media/images/images/image-settings-explained/ch-02-another-model.webp" }
  - { at: 116.6, title: "The cutout", image: "/media/images/images/image-settings-explained/ch-03-the-cutout.webp" }
thumb: images/image-settings-explained/thumb.webp
steps:
  - title: "Image Count"
    text: "Create › Images › GPT Image 2.5 Sunburst, Settings: drag Image Count to 1. The cost on Generate goes down, since you pay per image."
  - title: "Dimensions"
    text: "Open the Dimensions presets and pick 3:4 (1536x2048), a portrait frame for a standing character."
  - title: "Quality"
    text: "Step Quality to Medium: low to test ideas, medium for most work, high and above for finals and fine detail."
  - title: "Background"
    text: "Set Background to Transparent for a real cutout. It doesn't change the cost and works on GPT Image 2.5 Flare and Sunburst."
  - title: "Say it in the prompt too"
    text: "Swap the landing-platform sentence for one that asks for a cutout, or the model still draws a scene. Click Generate."
    prompt: |-
      Character art for VOLTARIS, an invented free-to-play hero shooter: the tank hero, a broad-shouldered bearded man with dark skin and a short black topknot, in riveted burnt-orange armor with cyan light strips. On the left side of the picture, a huge hexagonal energy shield glowing amber, nearly as tall as he is, held in his right hand with its bottom edge on the ground. He stands upright, full body, front three-quarter view, looking at the camera with a calm, confident expression. A cutout on a transparent background, no scenery, no floor, no shadow. The whole figure and the whole shield in frame with space around them, eye-level camera. No text, no logos, no watermark.

      Style: stylized 3D at the standard of the best free-to-play hero shooters: bold readable silhouettes, appealing expressive faces, hand-tuned stylized PBR (painted metal, soft cloth, rubber), saturated but controlled palette, cinematic key light with a cool rim.
  - title: "Switch to another model"
    text: "Pick Ideogram 4.5: Aspect Ratio 3:4 and Resolution 2K replace Dimensions, Quality runs from very low to high, there's no Background, and Magic Prompt goes Off to keep your words."
  - title: "Seed"
    text: "Leave Seed empty for a new start each run, or type a number to keep it fixed and change one thing at a time."
  - title: "Check the cutout"
    text: "Open the cutout from step 5. Info should read Image Count 1, Width 1,536, Height 2,048, Quality Medium, Background Transparent; zoom in on the topknot and the shield rim for clean edges."
agent_steps:
  - title: "Connect the Scenario MCP"
    text: "Connect your agent and sign in once at [mcp.scenario.com](https://mcp.scenario.com). Then tell it which Scenario team and project to use, so every file lands there."
  - title: "Give the agent the files"
    text: "Nothing to upload: the cutout is made from the prompt alone, with no reference."
  - title: "Brief the agent: the cutout"
    text: "The setting and the prompt both ask for the cutout."
    prompt: |-
      Use GPT Image 2.5 Sunburst (model_openai-gpt-image-2-5-sunburst): width 1536, height 2048, quality "medium", background "transparent", numOutputs 1. No reference. Quote the cost with a dry run first and wait for my go.

      Prompt, word for word:
      Character art for VOLTARIS, an invented free-to-play hero shooter: the tank hero, a broad-shouldered bearded man with dark skin and a short black topknot, in riveted burnt-orange armor with cyan light strips. On the left side of the picture, a huge hexagonal energy shield glowing amber, nearly as tall as he is, held in his right hand with its bottom edge on the ground. He stands upright, full body, front three-quarter view, looking at the camera with a calm, confident expression. A cutout on a transparent background, no scenery, no floor, no shadow. The whole figure and the whole shield in frame with space around them, eye-level camera. No text, no logos, no watermark.

      Style: stylized 3D at the standard of the best free-to-play hero shooters: bold readable silhouettes, appealing expressive faces, hand-tuned stylized PBR (painted metal, soft cloth, rubber), saturated but controlled palette, cinematic key light with a cool rim.
  - title: "Compare another model's settings"
    text: "Reading a schema is free. It shows what the panel would show for Ideogram 4.5: no background, an aspect ratio and resolution instead of pixels, Magic Prompt and Seed."
    prompt: |-
      Read the parameters of Ideogram 4.5 (model_ideogram-v4-5) with model_schema_get and list them next to those of GPT Image 2.5 Sunburst, with which ones move the cost. Don't run anything.
  - title: "Check the cutout"
    text: "Accept when it's the same character as the full scene, the whole figure and shield are in frame, and the edges hold on the topknot and the glowing shield rim. Reject a floor, shadow or scene left behind, cut-off parts or halos."
    prompt: |-
      Download the cutout and check that the background is really transparent and that there's no floor, shadow or scenery. Zoom on the topknot and the shield rim and tell me if the edges show halos. If they do, tell me before you run anything else.
  - title: "File the result and report"
    prompt: |-
      Add the cutout to a collection called "VOLTARIS tank hero" and report the model, the settings, its asset ID and the cost spent.
updated: 2026-11-11
canonical: https://learn.scenario.com/images/image-settings-explained
---

Last updated: 2026-11-11

# Image settings explained

> What each image setting does, its usual range, and which current models expose it, as of October 2026.

In short:

- Each image setting, its API parameter, its usual range, and which current models expose it.
- The panel shows only what the selected model accepts; its Parameters tab is the live source.
- Count, size, resolution and quality change the price; background, seed and magic prompt do not.


The settings under the prompt decide how many images you get, their size, how much effort goes into each one and whether they have a background. This lesson changes them one at a time on **GPT Image 2.5 Sunburst** to make a transparent cutout of the tank hero from [Generate images](https://learn.scenario.com/images/generate-images.md), then switches to **Ideogram 4.5** to show how the settings change with the model.

:::note As of November 2026
The models, settings and prices in this article are the ones Scenario offered in November 2026. New models and updates arrive often, so check the model library for newer options before you choose.
:::

:::steps Quick version
1. Pick the model first: the panel shows only the settings it accepts.
2. Set **Image Count**, then the size (**Dimensions**, or **Aspect Ratio** and **Resolution**).
3. Set **Quality**: low to test, high for finals.
4. For a cutout, set **Background** to **Transparent** and say it in the prompt too.
5. On models that have them, turn **Magic Prompt** off to keep your words, and set a **Seed** to repeat a result.
:::

## Before you start

- **Where they are.** Open **Create** › **Images**. The settings sit in the **Settings** section, under the prompt and **Reference Images**, just above **Generate**.
- **The model decides.** Each model accepts different settings, so the section changes when you change the model. What you see in the panel is always what the selected model accepts.
- **Some settings change the cost.** The cost shows on the **Generate** button and updates as you change settings. Count, size and quality move it; **Background** does not.
- **Change one thing at a time.** If you change three settings and the image gets better, you won't know which one did it.

This lesson starts from the tank hero prompt and settings of the previous lesson: GPT Image 2.5 Sunburst, two images, 2048 × 1152, quality high.

## Step 1: Image Count

**Image Count** is how many images one run makes. Drag the slider to **1**. The cost on **Generate** goes down, since you pay for each image.

Make several when you're exploring and want takes to compare. Make one when you know what you want, or when you test one setting change.

![Settings on GPT Image 2.5 Sunburst with Image Count at 1](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/9dcb936a7a99641fbf1e78cb37d4d9d6a9e00516de4fec00ff3be37a5d988b75.webp)

## Step 2: Dimensions

**Dimensions** sets the size of the image in pixels, width × height. Click the arrows next to it to open the presets. Each model offers its own sizes; on GPT Image 2.5 Sunburst they go from tall (9:21, 9:16) through square (1:1) to wide (16:9, 21:9).

Pick **3:4 (1536x2048)**, a portrait frame that fits a standing character.

![The Dimensions presets open, from 9:21 (1440x3360) down to 21:9](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/2ff6c51e30635fa2b422faff52752c2873b24e05c8c465bebca558720773b13b.webp)

Choose the shape for where the image will go: tall for a phone screen or a full-body character, square for an icon or a post, wide for a banner or a game background. Bigger sizes cost more and take longer. On GPT Image, both sides must be multiples of 16, up to 3840 pixels, and sizes above 2560 × 1440 are experimental, so the presets are the safe choice.

## Step 3: Quality

**Quality** is how much effort the model spends on each image. Click it to step through the levels, from low up to max.

- **Auto**, the default, lets the model decide.
- **Low** is quick. Use it to test ideas and prompts.
- **Medium** covers most work.
- **High** and above are for finals, and for images with small text, faces or fine detail.

The cost follows the quality. Set it to **Medium**.

![Quality set to Medium, with Dimensions at 1536 × 2048](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/ea74022ac162e4d024d18fb5e47df83d79fea545af67ee1f1dfb9b94ca08a4eb.webp)

## Step 4: Background

**Background** chooses between a full image and a cutout. It has three values:

- **Auto** lets the model decide.
- **Opaque** always draws a solid background.
- **Transparent** draws the subject alone, with a real transparent background, ready to place on anything.

Pick **Transparent**. This one doesn't change the cost.

![Background set to Transparent](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/bc86ec4acc9e39cd6440a3d9775f5b8e0a794c0880d0381e54d8ef3c3a0ee52c.webp)

Transparent backgrounds work on GPT Image 2.5 Flare and GPT Image 2.5 Sunburst. With most other models, remove the background as a separate step: see [Remove and replace backgrounds](https://learn.scenario.com/images/remove-and-replace-backgrounds.md).

## Step 5: Say it in the prompt too

The setting alone isn't enough: if the prompt describes a scene, the model still draws it. Swap the sentence about the landing platform for one that asks for a cutout. This is the prompt the film runs:

```
Character art for VOLTARIS, an invented free-to-play hero shooter: the tank hero, a broad-shouldered bearded man with dark skin and a short black topknot, in riveted burnt-orange armor with cyan light strips. On the left side of the picture, a huge hexagonal energy shield glowing amber, nearly as tall as he is, held in his right hand with its bottom edge on the ground. He stands upright, full body, front three-quarter view, looking at the camera with a calm, confident expression. A cutout on a transparent background, no scenery, no floor, no shadow. The whole figure and the whole shield in frame with space around them, eye-level camera. No text, no logos, no watermark.

Style: stylized 3D at the standard of the best free-to-play hero shooters: bold readable silhouettes, appealing expressive faces, hand-tuned stylized PBR (painted metal, soft cloth, rubber), saturated but controlled palette, cinematic key light with a cool rim.
```

![The prompt with the cutout sentence, and the settings: 1 image, 1536 × 2048, Medium, Transparent](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/df2863cfbd84ce267b5ecbb450ec605742b8fdd4d49e218f9df2f65a2a8bdeaa.webp)

Click **Generate**.

## Step 6: Switch to another model

While it runs, look at how another model's settings differ. Click the model name, type "Ideogram" in the search box and pick **Ideogram 4.5**.

![The model list searched for Ideogram, with Ideogram 4.5 at the top](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/136f2d9c4d03246fa910bbf70d1202961c15404cd15067147653d45abd8d7d83.webp)

Watch the **Settings** section change:

- Instead of pixels, this model takes an **Aspect Ratio** for the shape and a **Resolution**, **1K** or **2K**, for the size. Set the ratio to **3:4**. The resolution applies once you choose a ratio; on **Auto**, the model picks the shape.
- Its **Quality** runs from very low to high. Very low needs a reference image, so use it for edits, not to generate from a prompt alone.
- There is no **Background** setting, because this model can't make cutouts.
- **Magic Prompt** lets the model rewrite your prompt before it draws. It helps a short prompt. Turn it **Off** to keep your words exactly as written, as you should for a detailed prompt or exact text.
- There's also a **Copyright Detection** switch, which the film leaves off.

![Ideogram 4.5 settings: Aspect Ratio 3:4, Resolution 2K, Quality High, Magic Prompt Off, Image Count, Seed and Copyright Detection](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/730eb8f36d662309e691bd083112248f23e239bd462686164f7d50546f583828.webp)

## Step 7: Seed

**Seed** is the starting point of the image. Leave it empty to get a new one each run, or type a number to keep it fixed.

With the same model, prompt, settings and seed, you usually get the same image back. That makes the seed the tool for changing one thing at a time: fix it, change one word or one setting, and compare.

![A seed of 4242 typed in the Seed field](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/83d2b92b3e23c0a67c0e7377e7fa7ffc50abc6112c21ffa163e8c6cc6f2d9c94.webp)

## Step 8: Check the cutout

Back in your session, the cutout from Step 5 is ready. Click it to open it. The **Info** panel confirms the settings: Image Count 1, Width 1,536, Height 2,048, Quality Medium, Background Transparent. It's the same character, with no floor, no shadow and nothing behind him, ready to drop into a menu or a poster.

![The cutout open in the viewer, on a transparent background, with its settings in the Info panel](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f25ddfa119a556ba6b49178faad2f0290efcd0215447e3ee1c238842db9e5e99.webp)

:::check
**Accept when:** the character matches the one from the full scene, the whole figure and the shield are in frame, and the edges are clean on fine detail (the topknot, the glowing shield rim) when you zoom in.
**Reject when:** a floor, a shadow or part of a scene is still there (check the prompt still describes one), parts are cut off, or the edges show halos. Fix the prompt, or remove the background with a dedicated tool instead.
:::

## Settings reference

| Setting | What it does | Changes the cost | Seen on |
|---|---|---|---|
| **Image Count** | Images per run | Yes | Every model (GPT Image up to 10, Ideogram 4.5 up to 8, Nano Banana 2.1 up to 4) |
| **Dimensions** | Width × height in pixels, from presets or typed | Yes | GPT Image 2.5 Flare and Sunburst |
| **Aspect Ratio** | The shape, as a ratio such as 3:4 or 16:9 | Through the size it gives | Ideogram 4.5, Nano Banana 2.1 |
| **Resolution** | The size for that shape: 1K or 2K (Ideogram 4.5); 1K, 2K or 4K (Nano Banana 2.1) | Yes | Ideogram 4.5, Nano Banana 2.1 |
| **Quality** | Effort per image | Yes | GPT Image (low to max), Ideogram 4.5 (very low to high) |
| **Background** | **Auto**, **Opaque** or **Transparent** | No | GPT Image 2.5 Flare and Sunburst |
| **Magic Prompt** | Lets the model rewrite the prompt first: auto, on or off | No | Ideogram 4.5 |
| **Seed** | Fixes the starting point so a result can be repeated | No | Ideogram 4.5 and many others |
| Thinking level | How long the model plans before drawing: minimal, medium or high | Not covered here | Nano Banana 2.1 |

Some models, mostly Scenario's own and trained ones, also show **Sampling Steps** (how many passes the model makes; more is sharper and slower), **Guidance** (how strictly it follows the prompt) and, with a trained style, a **LoRA Scale**. Start those at their defaults and change them in small steps.

## Common problems

| What you see | What to do |
|---|---|
| A setting you expected isn't there | The selected model doesn't accept it. Switch to a model that does, for example GPT Image 2.5 Sunburst for **Background**. |
| The cutout still has a floor or a scene | Remove the scene from the prompt and say "a cutout on a transparent background, no scenery, no floor, no shadow". |
| The cutout's edges have halos on fine detail | Zoom in and check hair, smoke and thin glowing edges. If they don't hold up, use a background removal model instead. |
| The image doesn't follow your exact words | Turn **Magic Prompt** off. |
| The same settings give a different image every time | Leave nothing to chance: set a **Seed** and keep the model, prompt and settings the same. |
| The cost is higher than you expected | Lower **Image Count**, the size or **Quality**. Test at low quality, then run the final once. |
| Ideogram rejects very low quality | Very low needs a reference image. Use low or above to generate from a prompt alone. |

## What's next

You know what each setting does; now choose the model. The [Image models guide](https://learn.scenario.com/images/image-models-guide.md) compares the main image models and what each is best at. To place your cutout in a new scene, see [Remove and replace backgrounds](https://learn.scenario.com/images/remove-and-replace-backgrounds.md).

Source: https://learn.scenario.com/images/image-settings-explained