FLUX 3 Image API
Roko serves FLUX 3 Klein through Kie AI under flux-3-image. Omit reference images to generate from text, or supply images to select editing automatically. Choose 1K, 2K or 4K and request 1–4 outputs.
- Public model ID
flux-3-image- Output tiers
1K / 2K / 4K (1K default)- Billing
Per generated image by resolution- Endpoint
POST /v1/images/tasks- Catalog snapshot
What is FLUX 3 Image?
The public ID flux-3-image currently uses Kie AI’s FLUX 3 Klein integration. It does not expose every feature of the full BFL FLUX 3 Image API.
Keep the same model ID: requests without references generate images; requests with 1–10 public image URLs select editing automatically.
Choose 1K, 2K or 4K and 1–4 outputs. Submit an asynchronous task and poll for its result; billing uses the chosen resolution and successful output count.
Core capabilities
Text to image
Describe the subject, background and style without selecting a separate upstream model.
Reference editing
Supply 1–10 references and explain what to change and retain. Review the result manually.
Batch outputs
Set n from 1 to 4. Settlement uses the successful output count.
Three resolution tiers
Choose 1K, 2K or 4K with a supported aspect ratio.
FLUX 3 Image API pricing
Pricing comes from the current Roko catalog and is billed per generated image at the chosen resolution. The numbers matched Kie AI's standard price when reviewed, but upstream pricing and promotions do not determine Roko billing; confirm the Roko estimate and final settlement.
| Resolution | Price | Billing unit |
|---|---|---|
| 1k | $0.01 | per output |
| 2k | $0.016 | per output |
| 4k | $0.036 | per output |
Cost examples
Default tier — standard cost for web and social assets.
Settlement uses successful output count. This calculation uses the current price snapshot.
Parameters and supported scope
These fields reflect the current Kie integration. References require public HTTP / HTTPS URLs. Dedicated bounding-box layout, grounding and safety_tolerance controls are not exposed.
| Parameter | Values | Notes |
|---|---|---|
prompt | textarea | Describe the image to create or the changes to apply to references. · Required |
images | url · 1–10 | Use up to 10 public HTTP / HTTPS image URLs. Base64 and browser blob: URLs are not accepted. · Optional |
aspect_ratio | auto / 1:1 / 16:9 / 9:16 / 4:3 / 3:4 / 3:2 / 2:3 / 21:9 / 9:21 | For edits, auto follows the reference image. Text-to-image uses 1:1 when the ratio is omitted or set to auto. · Default: auto · Optional |
resolution | 1k / 2k / 4k | Selects an output resolution tier configured for this model; the value may directly affect price. · Default: 1k · Optional |
n | 1–4 | Number of images generated per request; more outputs cost more. · Default: 1 · Optional |
output_format | png / jpeg / webp | File format of generated images: png or jpeg. · Default: png · Optional |
enhance_prompt | boolean | Optional |
seed | integer | Optional |
Call FLUX 3 Image with Roko API
Run the async task flow: submit a task to get a task_id, then poll for the result. Never expose an API key in client-side code.
Submit generation
curl https://api.rokoapi.com/v1/images/tasks \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: YOUR_UNIQUE_REQUEST_ID" \
-d '{
"model": "flux-3-image",
"prompt": "A ceramic mug on a white studio background, soft side lighting.",
"aspect_ratio": "16:9",
"resolution": "2k"
}'Query recovery task
curl https://api.rokoapi.com/v1/tasks/$TASK_ID \
-H "Authorization: Bearer $API_KEY"POST /v1/images/tasks returns 202 with a task_id; once generation completes, query GET /v1/tasks/{task_id} for the result or failure reason. Idempotency-Key must be unique; replays with the same key return the original task without re-billing.
Best-fit scenarios
Product scene drafts
Describe the product, lighting and background. Check composition at 1K first.
A ceramic mug on a white studio background, soft side lighting.
Background and color changes
Provide references and clear edit instructions. Do not assume all other pixels will remain unchanged.
Change the mug to blue while keeping its shape and the white background.
Multi-reference compositions
Explain which subject, style or setting each reference contributes.
Use the product from image 1 and the background from image 2.
Alternative drafts
Request up to four outputs with n, then review and select a version.
Create alternative studio compositions of a ceramic mug.
Important limitations
- Supply 1–10 public HTTP / HTTPS reference URLs; Base64, Data URLs and blob: URLs are not accepted.
- Nine fixed aspect ratios are supported. Editing accepts auto; text-to-image uses 1:1 when auto is supplied or the ratio is omitted.
- This is the Klein integration. It does not guarantee exact preservation outside an edited region or expose the full API’s dedicated layout and web-search controls.
- Submission only acknowledges a task. Poll to a terminal state and review the output before use.
FLUX 3 Image or another image model?
Choose using the task, reference limits, resolution and current catalog prices. This page does not report controlled quality or speed comparisons.
All models →FLUX 3 Image FAQ
What model ID should I use for FLUX 3 Image?
Use the public Roko API model ID flux-3-image exactly as shown in the model field.
How much does the FLUX 3 Image API cost?
Billing is per generated image at the chosen resolution: $0.008 at 1K, $0.0125 at 2K, $0.0325 at 4K. The values matched Kie AI's standard price when reviewed, but Roko billing follows the project estimate and final settlement at submission.
Do generation and editing need different model IDs?
No. Keep flux-3-image. Without references it generates from text; references automatically select editing.
How do I supply references?
Send 1–10 public HTTP / HTTPS image URLs in image_urls or images. image and input_reference aliases are also accepted.
Which resolutions and aspect ratios are supported?
Use 1K, 2K or 4K. Fixed ratios are 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9 and 9:21. Editing accepts auto; text-to-image auto uses 1:1.
Which endpoint does FLUX 3 Image use?
Submit async tasks via POST /v1/images/tasks to get a task_id, then poll GET /v1/tasks/{task_id} for the result or failure reason.
Is this the full BFL API?
This route serves FLUX 3 Klein through Kie AI. Dedicated bounding-box layout and live web-search parameters are not exposed; do not infer the full API’s capabilities.
How many images can one task request?
Set n to 1–4; the default is 1. Billing uses successful output count and the selected resolution.