# FLUX.1 Kontext [dev]

> Edit an existing image from a text instruction — change objects, style, background, or text — while keeping the rest of the photo consistent.

## Overview

- **Endpoint**: `https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev`
- **Model ID**: `black-forest-labs/flux-kontext/dev`
- **Category**: image-to-image
- **Kind**: inference
- **Tags**: flux, flux-kontext, kontext, black-forest-labs, image-to-image, image-editing, instruction-edit, edit

## Pricing

- **Price**: $0.025 per megapixel

## Request Lifecycle

This model runs on the ModelRunner **asynchronous queue API** — a single POST does not return the output. Every call requires an `Authorization: Key $MODEL_RUNNER_KEY` header. Run three steps:

1. **Submit** — `POST https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev` with a JSON body holding the input fields at the top level. The body may also include a reserved top-level `metadata` object — a flat string map (max 16 keys, key ≤64 / value ≤512 chars) stored on the request for your own tagging. It is never sent to the model; filter your request history with `GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>` (exact key=value matches, AND-ed). The response carries request handles only (no output yet):

```json
{
  "status": "IN_QUEUE",
  "request_id": "<21-char id>",
  "status_url": "https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev/requests/<request_id>/status",
  "response_url": "https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev/requests/<request_id>",
  "cancel_url": "https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev/requests/<request_id>/cancel"
}
```

2. **Poll status** — `GET <status_url>` until `status` is `COMPLETED`. Possible values are `IN_QUEUE`, `IN_PROGRESS`, `COMPLETED`, `FAILED`, `CANCELLED`. A `FAILED` request responds with HTTP 400 and an `error` field.

3. **Read result** — `GET <response_url>`. Returns the finished request, including the generated `output`:

```json
{ "id": "<request_id>", "status": "COMPLETED", "output": ..., "input": ... }
```

The JavaScript and Python SDKs below perform steps 2–3 for you. In any language without an SDK (Swift, Go, Kotlin, etc.) you must implement the polling loop and the final result fetch yourself — see the cURL example for the full flow.

### Input Schema

- **`seed`** (`integer | null`, _optional_):
  Random seed. The same seed, prompt, and image produce the same result.

- **`prompt`** (`string`, _required_):
  The edit instruction. Describe the change to apply to the source image (e.g. 'change the wooden floor to white marble') and name anything that should stay the same.

- **`image_url`** (`string`, _required_):
  The source image to edit. Unmentioned regions are kept consistent while the described change is applied.

- **`num_images`** (`integer`, _optional_):
  The number of images to generate. Each generated image is billed.
  - Default: `1`
  - Range: `1` to `4`

- **`acceleration`** (`AccelerationEnum`, _optional_):
  Inference acceleration level. 'none' is highest quality; 'regular' and 'high' are progressively faster with a small quality trade-off. Does not change the price.
  - Default: `"none"`
  - Options: `"none"`, `"regular"`, `"high"`

- **`output_format`** (`OutputFormatEnum`, _optional_):
  The format of the generated image.
  - Default: `"jpeg"`
  - Options: `"jpeg"`, `"png"`

- **`guidance_scale`** (`number`, _optional_):
  How strongly the result follows the edit instruction. Raise for stronger adherence; lower if the edit looks over-applied.
  - Default: `2.5`
  - Range: `1` to `20`

- **`resolution_mode`** (`ResolutionModeEnum`, _optional_):
  Output resolution. 'match_input' keeps the source image's resolution (recommended for alignment); 'auto' lets the model pick an optimal resolution matching the input aspect; or force an aspect ratio such as '16:9', '4:3', or '9:16'.
  - Default: `"match_input"`
  - Options: `"auto"`, `"match_input"`, `"1:1"`, `"16:9"`, `"21:9"`, `"3:2"`, `"2:3"`, `"4:5"`, `"5:4"`, `"3:4"`, `"4:3"`, `"9:16"`, `"9:21"`

- **`num_inference_steps`** (`integer`, _optional_):
  Number of denoising steps. More steps can improve detail at the cost of speed.
  - Default: `28`
  - Range: `10` to `50`

- **`enable_safety_checker`** (`boolean`, _optional_):
  If set to true, the safety checker will be enabled.
  - Default: `true`

### Output Schema

_No `Output` schema properties are available._

## Default Example

**Input**

```json
{
  "prompt": "Repaint the car in glossy metallic teal and change the scene to a rainy neon-lit city street at night, reflections on the wet cobblestones.",
  "image_url": "https://media.modelrunner.ai/I0oucXokx5BIPjZDjvig6.png",
  "acceleration": "none",
  "output_format": "jpeg",
  "guidance_scale": 2.5,
  "resolution_mode": "match_input",
  "num_inference_steps": 28,
  "enable_safety_checker": true
}
```

**Output**

```json
[
  "https://media.modelrunner.ai/fQZGZfsnCcsI8v0cR7RTJ.jpeg"
]
```
## Usage Examples

### cURL

The queue API is asynchronous: submit the request, poll `status_url` until it is `COMPLETED`, then read the result from `response_url`. Requires `jq`.

```bash
# 1. Submit the request (returns request handles, not the output)
SUBMIT=$(curl --silent --request POST \
  --url https://queue.modelrunner.run/black-forest-labs/flux-kontext/dev \
  --header "Authorization: Key $MODEL_RUNNER_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "prompt": "Repaint the car in glossy metallic teal and change the scene to a rainy neon-lit city street at night, reflections on the wet cobblestones.",
  "image_url": "https://media.modelrunner.ai/I0oucXokx5BIPjZDjvig6.png",
  "acceleration": "none",
  "output_format": "jpeg",
  "guidance_scale": 2.5,
  "resolution_mode": "match_input",
  "num_inference_steps": 28,
  "enable_safety_checker": true
}')

STATUS_URL=$(echo "$SUBMIT" | jq -r '.status_url')
RESPONSE_URL=$(echo "$SUBMIT" | jq -r '.response_url')

# 2. Poll until the request leaves the queue / in-progress state
while true; do
  STATUS=$(curl --silent --url "$STATUS_URL" \
    --header "Authorization: Key $MODEL_RUNNER_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  case "$STATUS" in
    COMPLETED) break ;;
    FAILED|CANCELLED) echo "Request $STATUS"; exit 1 ;;
  esac
  sleep 1
done

# 3. Read the finished request, including the generated output
curl --silent --url "$RESPONSE_URL" \
  --header "Authorization: Key $MODEL_RUNNER_KEY"
```

### JavaScript

```javascript
import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("black-forest-labs/flux-kontext/dev", {
  input: {
  "prompt": "Repaint the car in glossy metallic teal and change the scene to a rainy neon-lit city street at night, reflections on the wet cobblestones.",
  "image_url": "https://media.modelrunner.ai/I0oucXokx5BIPjZDjvig6.png",
  "acceleration": "none",
  "output_format": "jpeg",
  "guidance_scale": 2.5,
  "resolution_mode": "match_input",
  "num_inference_steps": 28,
  "enable_safety_checker": true
}
});

console.log(result.data);
```

### Python

```python
import asyncio
import modelrunner_ai

async def main():
    response = await modelrunner_ai.submit_async(
        "black-forest-labs/flux-kontext/dev",
        arguments={
  "prompt": "Repaint the car in glossy metallic teal and change the scene to a rainy neon-lit city street at night, reflections on the wet cobblestones.",
  "image_url": "https://media.modelrunner.ai/I0oucXokx5BIPjZDjvig6.png",
  "acceleration": "none",
  "output_format": "jpeg",
  "guidance_scale": 2.5,
  "resolution_mode": "match_input",
  "num_inference_steps": 28,
  "enable_safety_checker": true
}
    )
    result = await response.get()
    print(result["output"])

asyncio.run(main())
```

## Additional Resources

- [Playground](https://modelrunner.ai/models/black-forest-labs/flux-kontext/dev)
- [OpenAPI Schema](https://modelrunner.ai/models/black-forest-labs/flux-kontext/dev/openapi.json)
- [LLM Instructions](https://modelrunner.ai/models/black-forest-labs/flux-kontext/dev/llms.txt)