Images
remix
New images steered by the look of pictures you already have.
<!-- Generated by apps/docs/scripts/gen-tools.mjs. Do not edit here: the words live in apps/web/src/lib/tools/krea-2.tool.json (agent, hint) and packages/shared/src/tools/setting-hints.ts. --> On the rail as **krea-2**, **remix**. ## When to use it Making a new image that should carry the look of work you have already made: drop up to eight references, write a prompt, and set how hard each one pulls on the result. From a picture remixes one picture into that look. ## When to reach for something else It blends the mood and style of the references rather than copying them exactly, so it will not reproduce one subject pose for pose. For a precise edit to one part of a picture, a different tool fits better. ## How to use it 1. Use the card's **Style references** area. Up to eight style references. Drag each one to set how hard it pulls on the result. 2. Write what you want in **Prompt**. 3. Pick a mode on the card: **From words**, **From a picture**. 4. Pick a model in the pill at the top of the card, or keep **Medium**. 5. Press **Run**. ## On the card | Control | What it does | Choices | | - | - | - | | Style references | Up to eight style references. Drag each one to set how hard it pulls on the result. | | | Prompt | What you want, in your own words. The card sends it with your inputs. | Your words | | Run | Starts the run. The number on it is what this run costs in credits, worked out before you press. | | | ×1 | Runs the card several times in a row with the same settings. | 1 to 99 | ## Modes ### From words A new picture from your prompt ### From a picture Remix a picture you choose ## Models The pill at the top of the card picks the model. **Medium** is the one a new card starts on. | Model | Good for | | - | - | | [Large](#large) | Krea 2 at the Large tier: text-to-image steered by up to ten dropped style references, each with its own drag-set strength, and the engine Digital Twin uses to redraw a dropped picture with a Resemblance slider. | | [Medium](#medium) | Krea 2 at the Medium tier, the default on the pill: the same style-reference steering as Large at half the cost, the setting for exploring directions before committing to a final render. | | [Turbo](#turbo) | Krea 2 at the Turbo tier, the fastest and cheapest of the three: the same style-reference steering for rapid rounds of ideas where turnaround beats finish. | ### Large **Not the one for:** Output is 1K only, the API has not shipped 2K. It re-composes freely from references rather than holding a subject in place, so it is not an editor, and attaching style references adds a small per-image surcharge. Behind the settings button: | Setting | What it does | Choices | | - | - | - | | Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. "Match input" copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 | | Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number | ### Medium **Not the one for:** Output is 1K only and the references steer the look rather than pin a subject in place, so it is not an editor. When the final render should come from the top quality tier, step up to Large. Behind the settings button: | Setting | What it does | Choices | | - | - | - | | Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. "Match input" copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 | | Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number | ### Turbo **Not the one for:** Output is 1K only and the references steer the look rather than pin a subject in place. It is the speed tier, so a hero render or a final comp belongs on Medium or Large. Behind the settings button: | Setting | What it does | Choices | | - | - | - | | Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. "Match input" copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 | | Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number |On the rail as krea-2, remix.
When to use it
Section titled “When to use it”Making a new image that should carry the look of work you have already made: drop up to eight references, write a prompt, and set how hard each one pulls on the result. From a picture remixes one picture into that look.
When to reach for something else
Section titled “When to reach for something else”It blends the mood and style of the references rather than copying them exactly, so it will not reproduce one subject pose for pose. For a precise edit to one part of a picture, a different tool fits better.
How to use it
Section titled “How to use it”- Use the card’s Style references area. Up to eight style references. Drag each one to set how hard it pulls on the result.
- Write what you want in Prompt.
- Pick a mode on the card: From words, From a picture.
- Pick a model in the pill at the top of the card, or keep Medium.
- Press Run.
On the card
Section titled “On the card”| Control | What it does | Choices |
|---|---|---|
| Style references | Up to eight style references. Drag each one to set how hard it pulls on the result. | |
| Prompt | What you want, in your own words. The card sends it with your inputs. | Your words |
| Run | Starts the run. The number on it is what this run costs in credits, worked out before you press. | |
| ×1 | Runs the card several times in a row with the same settings. | 1 to 99 |
From words
Section titled “From words”A new picture from your prompt
From a picture
Section titled “From a picture”Remix a picture you choose
Models
Section titled “Models”The pill at the top of the card picks the model. Medium is the one a new card starts on.
| Model | Good for |
|---|---|
| Large | Krea 2 at the Large tier: text-to-image steered by up to ten dropped style references, each with its own drag-set strength, and the engine Digital Twin uses to redraw a dropped picture with a Resemblance slider. |
| Medium | Krea 2 at the Medium tier, the default on the pill: the same style-reference steering as Large at half the cost, the setting for exploring directions before committing to a final render. |
| Turbo | Krea 2 at the Turbo tier, the fastest and cheapest of the three: the same style-reference steering for rapid rounds of ideas where turnaround beats finish. |
Not the one for: Output is 1K only, the API has not shipped 2K. It re-composes freely from references rather than holding a subject in place, so it is not an editor, and attaching style references adds a small per-image surcharge.
Behind the settings button:
| Setting | What it does | Choices |
|---|---|---|
| Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. “Match input” copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 |
| Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number |
Medium
Section titled “Medium”Not the one for: Output is 1K only and the references steer the look rather than pin a subject in place, so it is not an editor. When the final render should come from the top quality tier, step up to Large.
Behind the settings button:
| Setting | What it does | Choices |
|---|---|---|
| Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. “Match input” copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 |
| Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number |
Not the one for: Output is 1K only and the references steer the look rather than pin a subject in place. It is the speed tier, so a hero render or a final comp belongs on Medium or Large.
Behind the settings button:
| Setting | What it does | Choices |
|---|---|---|
| Aspect Ratio | The shape of the result, like 1:1 square, 16:9 wide, or 9:16 tall. “Match input” copies the source shape. | Auto, Square 1:1, Horizontal 4:3, Horizontal 3:2, Horizontal 16:9, Horizontal 2.35:1, Vertical 4:5, Vertical 3:4, Vertical 2:3, Vertical 9:16 |
| Seed | The starting dice roll. The same seed and settings give the same result every time; change it for a fresh variation, or leave it at -1 for a new random one. | A number |