Images
flux-kontext
Describe a change, keep the rest of the picture.
<!-- Generated by apps/docs/scripts/gen-tools.mjs. Do not edit here: the words live in apps/web/src/lib/tools/flux-kontext.tool.json (agent, hint) and packages/shared/src/tools/setting-hints.ts. --> On the rail as **flux-kontext**. ## When to use it Changing one thing in a picture by describing it in words, with no mask to paint and the rest of the frame left exactly where it was. ## When to reach for something else It edits one picture on one engine and cannot be pointed at an exact region, so an everyday edit in words, a comparison across models or a mix of pictures belongs in image-edit, and an exact, painted-in edit in inpainting. ## How to use it 1. Drop a picture onto the card's **Image** slot, from the board or your computer. 2. Write what you want in **Prompt**. 3. Press **Run**. ## On the card | Control | What it does | Choices | | - | - | - | | Image | The picture the tool works from. Drop one from the board or your computer. | One picture | | 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 | ## Settings These sit behind the settings button on the card. | Setting | What it does | Choices | | - | - | - | | Strength | How far the result can drift from the original image. Low stays close to the source; high lets it change a lot. | 0.01 to 1 | | Guidance Scale | How tightly the result sticks to your prompt. Higher follows your words more literally; lower gives the tool more freedom. | 1 to 20 | | Steps | How many passes the tool takes to refine the result. More looks a bit smoother but is slower. | 10 to 50 | | 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. | -1 to 999999 |On the rail as flux-kontext.
When to use it
Section titled “When to use it”Changing one thing in a picture by describing it in words, with no mask to paint and the rest of the frame left exactly where it was.
When to reach for something else
Section titled “When to reach for something else”It edits one picture on one engine and cannot be pointed at an exact region, so an everyday edit in words, a comparison across models or a mix of pictures belongs in image-edit, and an exact, painted-in edit in inpainting.
How to use it
Section titled “How to use it”- Drop a picture onto the card’s Image slot, from the board or your computer.
- Write what you want in Prompt.
- Press Run.
On the card
Section titled “On the card”| Control | What it does | Choices |
|---|---|---|
| Image | The picture the tool works from. Drop one from the board or your computer. | One picture |
| 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 |
Settings
Section titled “Settings”These sit behind the settings button on the card.
| Setting | What it does | Choices |
|---|---|---|
| Strength | How far the result can drift from the original image. Low stays close to the source; high lets it change a lot. | 0.01 to 1 |
| Guidance Scale | How tightly the result sticks to your prompt. Higher follows your words more literally; lower gives the tool more freedom. | 1 to 20 |
| Steps | How many passes the tool takes to refine the result. More looks a bit smoother but is slower. | 10 to 50 |
| 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. | -1 to 999999 |