FLUX API

Model Introduction

Exchange Token provides FLUX image generation through these API paths:

• POST /v1/flux-2-pro
• POST /v1/flux-2-max
• POST /v1/flux-3-image

Submit a generation request first, then poll the returned polling_url or GET /v1/get_result?id=....

Calling Method

API Endpoint

https://api.exchangetoken.ai

Authentication

FLUX endpoints use the following authentication header:

x-key: YOUR_API_KEY

You can also use Authorization: Bearer YOUR_API_KEY. Use the same API key to create the task and query its result.

Content Type

Content-Type: application/json

Task Submission

FLUX 2 pro/max Request

Use POST /v1/flux-2-pro or POST /v1/flux-2-max.

Request Fields

Field	Type	Required	Description
prompt	string	Yes	Text prompt for image generation or editing
input_image	string	No	First reference image as an accessible image URL or Base64; omit for text-to-image generation
input_image_2 to input_image_8	string	No	Reference images 2–8; one image per field
seed	integer	No	Random seed for reproducibility, such as 42
width	integer	No	Output width in pixels; must be at least 64 when specified
height	integer	No	Output height in pixels; must be at least 64 when specified
safety_tolerance	integer	No	The current API accepts 0–5, with 0 the strictest
output_format	string	No	Output image format; use jpeg or png

FLUX 3 Image Request

Use POST /v1/flux-3-image.

Request Fields

Field	Type	Required	Description
prompt	string	Yes	Text prompt for image generation or editing; must not be empty or contain only whitespace
images	string or string[]	No	1–10 reference images as HTTP(S) URLs or Base64; omit this field when there are no references, rather than sending null or an empty array
resolution	string	No	Resolution tier: 768sq, 1k, 1.5k, 2k, or 4k; defaults to 1k
aspect_ratio	string	No	Output aspect ratio; defaults to auto. See the values below
grounding	boolean	No	Whether to enable web and image search; false disables both
safety_tolerance	integer	No	The current API accepts 0–4, with 0 the strictest

Supported aspect_ratio values: 21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, 1:2, 9:21, and auto.

Response

Field	Type	Description
id	string	Gateway task ID for subsequent queries
polling_url	string	Full result URL; send a GET request with the same API key

Query Task Result

Request Fields

GET /v1/get_result?id=TASK_ID

Field	Location	Type	Required	Description
id	Query	string	Yes	Task ID from the submit response

Response Fields

Field	Type	Description
id	string	Gateway task ID
status	string	Currently Pending, Ready, or Error
result	object or null	Contains sample on success; null while pending or on failure
result.sample	string	Complete generated image URL
progress	number	Optional progress value; omitted when unavailable
details	object[]	Optional failure details; each entry contains only msg
details[].msg	string	Failure reason

Platform Error Responses

When a request fails validation by the Exchange Token platform or Exchange Token cannot complete request processing, the response uses an HTTP 4xx or 5xx status code with the following structure:

Field	Type	Description
error	object	Platform error information
error.type	string	Error type, such as invalid_request or rate_limited
error.message	string	Detailed description of the error

Common HTTP status codes:

Status Code	Description
400	Invalid parameters or missing required parameters
401	Missing or invalid API key
402	Insufficient available quota
403	Access denied, for example because the API key is disabled, expired, or the client IP is not allowed
404	The task does not exist or does not belong to the current API key
429	Request rate or concurrency limit exceeded
500	Internal platform processing failure
502	Upstream communication failure or an invalid upstream response
503	Service temporarily unavailable, for example because no channel is available

Examples

Submit a Generation Task

FLUX 2 pro Text-to-Image

curl -X POST "https://api.exchangetoken.ai/v1/flux-2-pro" \
  -H "x-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cinematic portrait of a silver robot walking in neon rain",
    "width": 1024,
    "height": 1024,
    "seed": 42,
    "safety_tolerance": 2,
    "output_format": "png"
  }'

FLUX 2 max with Reference Images

Replace the example URLs with accessible image URLs.

curl -X POST "https://api.exchangetoken.ai/v1/flux-2-max" \
  -H "x-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Keep the teapot shape from the first image, apply the colors from the second image, and place it on a wooden table",
    "input_image": "https://example.com/teapot.png",
    "input_image_2": "https://example.com/color-reference.png",
    "width": 1024,
    "height": 1024,
    "output_format": "png"
  }'

FLUX 3 Image Text-to-Image

curl -X POST "https://api.exchangetoken.ai/v1/flux-3-image" \
  -H "x-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A cinematic portrait of a silver robot walking in neon rain",
    "resolution": "2k",
    "aspect_ratio": "1:1",
    "grounding": false,
    "safety_tolerance": 2
  }'

FLUX 3 Image with Reference Images

Replace the example URLs with accessible image URLs.

curl -X POST "https://api.exchangetoken.ai/v1/flux-3-image" \
  -H "x-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Keep the teapot shape from the first image, apply the colors from the second image, and place it on a wooden table",
    "images": [
      "https://example.com/teapot.png",
      "https://example.com/color-reference.png"
    ],
    "resolution": "1k",
    "aspect_ratio": "4:3"
  }'

Response

{
  "id": "sg_img_123456",
  "polling_url": "https://api.exchangetoken.ai/v1/get_result?id=sg_img_123456"
}

Query Task Result

Replace id with the task ID from the submit response.

curl "https://api.exchangetoken.ai/v1/get_result?id=sg_img_123456" \
  -H "x-key: YOUR_API_KEY"

Poll the same task ID until status changes to Ready or Error.

Pending Response Example

{
  "id": "sg_img_123456",
  "status": "Pending",
  "result": null,
  "progress": 0.45
}

Ready Response Example

{
  "id": "sg_img_123456",
  "status": "Ready",
  "result": {
    "sample": "https://cdn.example.com/generated-image.png?signature=example"
  }
}

Read result.sample to retrieve the image. Preserve the complete URL, including its query parameters, when downloading it.

Error Responses

Platform Parameter Error

A FLUX 3 request with an unsupported resolution returns HTTP 400:

{
  "error": {
    "type": "invalid_request",
    "message": "FLUX 3 resolution must be one of 768sq, 1k, 1.5k, 2k, 4k"
  }
}

Upstream Task Error

Querying a failed task returns HTTP 200, with the upstream error message in details[0].msg:

{
  "id": "sg_img_123456",
  "status": "Error",
  "result": null,
  "details": [
    {
      "msg": "image generation failed"
    }
  ]
}