Black Forest LabsTutorial
How to Use FLUX.2 [pro] API
A step-by-step FLUX.2 [pro] API tutorial with Python, Node.js and cURL: submit a text to image job, poll it or receive a webhook, and download the image. The example below costs $0.026.
- Endpoints
- 2
- Text to Image · Image Edit
- Example request
- $0.026
- Text to Image
- Languages
- 3
- Python · Node.js · cURL
- Failed runs
- $0
- refunded automatically
Get an API key
Create an account, add credits (from $10) and create a key in Dashboard → API keys. Store it as CHEAPERGEN_API_KEY on your server. Every request sends it as a bearer token.
Choose a FLUX.2 [pro] endpoint
Each task has its own endpoint. All of them take a JSON body and return a generation you can poll.
| Task | Endpoint | Required | From |
|---|---|---|---|
| Text to Image | bfl/flux-2-pro/text-to-image | prompt | $0.0264/image |
| Image Edit | bfl/flux-2-pro/edit | prompt, image_urls | $0.0264/image |
Submit a text to image job
This program submits the job, polls every 3 seconds until it finishes and prints the image URL. Swap the prompt and settings for your own.
import os, time, requests
API = "https://cheapergen.com/api/v1"
headers = {"Authorization": f"Bearer {os.environ['CHEAPERGEN_API_KEY']}"}
# 1. Submit the job
job = requests.post(f"{API}/run/bfl/flux-2-pro/text-to-image", headers=headers, json={
"prompt": "Editorial photo of a ceramic vase with dried flowers on a linen-covered table, soft north light, film grain",
"aspect_ratio": "1:1",
"resolution": "1MP"
}).json()
# 2. Poll until it finishes (or pass "webhook_url" and skip polling)
while job["status"] in ("queued", "running"):
time.sleep(3)
job = requests.get(f"{API}/generations/{job['id']}", headers=headers).json()
print(job["status"], job["output"]["images"][0]["url"] if job["status"] == "succeeded" else job["error"])Set the parameters that matter
The settings for bfl/flux-2-pro/text-to-image, with what each one does to the price.
| Parameter | Type · default | What it does | Price effect |
|---|---|---|---|
| aspect_ratio | string (enum) · 1:1 | Output image shape. 1:1 · 4:3 · 3:4 · 16:9 · 9:16 · 3:2 · 2:3 · 21:9 · 9:21 | Yes |
| resolution | string (enum) · 1MP | Output size in megapixels (1 MP ≈ 1024×1024). FLUX.2 is billed per megapixel. 1MP · 2MP · 3MP · 4MP | 1 MP: $0.0264/image 2 MP: $0.0396/image 3 MP: $0.0385/image 4 MP: $0.066/image |
| prompt_upsampling | boolean · false | Let the model expand and rewrite the prompt before generating. | None |
| seed | integer | Use the same seed and settings to reproduce a result. Leave empty for random. | None |
Other FLUX.2 [pro] endpoints take the same shape with their own inputs:
POST /api/v1/run/bfl/flux-2-pro/edit
{
"prompt": "Place the sneaker from image 1 on the concrete plinth in image 2, matching the hard afternoon shadows",
"image_urls": [
"https://cdn.cheapergen.com/samples/portrait.jpg"
],
"aspect_ratio": "auto",
"resolution": "1MP"
}Get the result
A job moves from queued to running to succeeded, failed or canceled. Poll GET /api/v1/generations/{id}, or pass webhook_url and we POST the result to you. Download the output and store it yourself.
200 — succeeded
{
"id": "gen_8c2QmX1nq0k4",
"object": "generation",
"model": "bfl/flux-2-pro",
"endpoint": "bfl/flux-2-pro/text-to-image",
"mode": "text-to-image",
"status": "succeeded",
"input": {
"prompt": "Editorial photo of a ceramic vase with dried flowers on a linen-covered table, soft north light, film grain",
"aspect_ratio": "1:1",
"resolution": "1MP"
},
"output": {
"images": [
{
"url": "https://cdn.cheapergen.com/outputs/gen_8c2Q-0.png",
"content_type": "image/png",
"width": 2048,
"height": 2048
}
]
},
"error": null,
"cost": {
"amount_usd": 0.2255,
"currency": "usd",
"estimated": false,
"official_usd": 0.264,
"saved_usd": 0.0385
},
"source": "api",
"created_at": "2026-10-15T09:21:04Z",
"started_at": "2026-10-15T09:21:06Z",
"completed_at": "2026-10-15T09:22:31Z"
}Price a request before you run it
The quote endpoint is free and needs no key. It returns the price, the official price and your discount for exactly the input you send.
curl -X POST https://cheapergen.com/api/v1/quote/bfl/flux-2-pro/text-to-image \
-H "Content-Type: application/json" \
-d '{"prompt":"Editorial photo of a ceramic vase with dried flowers on a linen-covered table, soft north light, film grain","aspect_ratio":"1:1","resolution":"1MP"}'| Status | Code | Meaning |
|---|---|---|
| 400 | invalid_request | A parameter is missing or out of range — the message names it. |
| 401 | unauthorized | Missing, malformed or revoked API key. |
| 402 | insufficient_balance | Not enough credits for the quoted price; top up or enable auto top-up. |
| 429 | rate_limited | Too many requests or generations in progress; back off and retry. |
| 502 | upstream_error | The model provider failed; the hold is refunded and you can retry. |
| 503 | model_unavailable | No healthy route right now; retry later or pick another model. |
FAQ
FLUX.2 [pro] API questions
No. The API is plain HTTPS with JSON, so the Python (requests) and Node.js (fetch) examples on this page are complete programs. Any language that can send an HTTP request works.
Yes. Add "webhook_url" to the request body and we POST the finished generation to it, signed with a CheaperGen-Signature header (HMAC-SHA256 of the timestamp and raw body). Polling GET /api/v1/generations/{id} works too.
Pass public URLs in "image_urls". Files without a public URL can be uploaded first with POST /api/v1/uploads, which returns a URL to use in the request.
Text to Image starts at $0.0264/image; the example request on this page costs $0.026, 12% below the official price. Failed generations are refunded automatically.