flux-2-pro

Quick example

curl https://reapi.ai/api/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-2-pro",
    "prompt": "a cinematic product photograph of an espresso machine",
    "resolution": "2MP",
    "size": "16:9"
  }'

import requests

resp = requests.post(
    "https://reapi.ai/api/v1/images/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "model": "flux-2-pro",
        "prompt": "a cinematic product photograph of an espresso machine",
        "resolution": "2MP",
        "size": "16:9",
    },
    timeout=30,
)
print(resp.json())

const resp = await fetch('https://reapi.ai/api/v1/images/generations', {
  method: 'POST',
  headers: {
    Authorization: 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    model: 'flux-2-pro',
    prompt: 'a cinematic product photograph of an espresso machine',
    resolution: '2MP',
    size: '16:9',
  }),
});

console.log(await resp.json());

package main

import (
	"bytes"
	"fmt"
	"io"
	"net/http"
)

func main() {
	body := []byte(`{
		"model": "flux-2-pro",
		"prompt": "a cinematic product photograph of an espresso machine",
		"resolution": "2MP",
		"size": "16:9"
	}`)

	req, _ := http.NewRequest("POST",
		"https://reapi.ai/api/v1/images/generations", bytes.NewReader(body))
	req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
	req.Header.Set("Content-Type", "application/json")

	resp, err := http.DefaultClient.Do(req)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()

	out, _ := io.ReadAll(resp.Body)
	fmt.Println(string(out))
}

Submit response

{
  "id": "task_019dfd44b7fd74168541552a3260a623",
  "model": "flux-2-pro",
  "status": "processing",
  "created_at": 1785323949
}

Poll GET /api/v1/tasks/{task_id} until status is completed or failed.

Authentication

Authorization: Bearer YOUR_API_KEY

Create a key from your dashboard.

Endpoint

POST https://reapi.ai/api/v1/images/generations

Submission is asynchronous — the call returns a task id immediately and the image is retrieved by polling the Tasks API.

Tiers

All three share one parameter set, except that steps and guidance are accepted only by flux-2-flex-std.

Modes

Mode is implicit in whether you send reference images:

Request body

model — string, required

One of flux-2-pro, flux-2-max, flux-2-flex-std.

prompt — string, required

Description of the image to generate, or the edit to apply to the reference images.

resolution — string, default 2MP

Output resolution tier: 1MP, 2MP, 3MP, 4MP. Applies when size is an aspect ratio.

Legacy aliases are accepted and map onto the four tiers: 512 / 512P / 1M → 1MP, 1K / 1024 → 2MP, 2K / 2048 → 3MP, 4K → 4MP.

size — string, default 1:1

Output aspect ratio: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9, 9:21.

A pixel string such as 1024x1536 is also accepted and takes priority over resolution.

width / height — integer, optional

Exact output dimensions. Must be sent as a pair — one without the other is a 400. Each side is at least 64 pixels, and width × height must not exceed 4 MP (4,194,304 pixels).

A complete pair has the highest priority and overrides both resolution and size.

image_urls — array, optional

Up to 8 public http(s) image URLs.

output_format — string, default jpeg

jpeg, png, or webp.

n — integer, default 1

1 is the only supported value; submit concurrent requests for more.

seed — integer, optional

Fix the seed to make a generation reproducible.

prompt_upsampling — boolean, default false

When true, the model rewrites and expands your prompt before generating.

safety_tolerance — integer, default 2

Moderation strictness from 0 (strictest) to 5 (most permissive).

steps — integer, optional — flux-2-flex-std only

Sampling steps, 1–50. Sending this to flux-2-pro or flux-2-max returns a 400.

guidance — number, optional — flux-2-flex-std only

Prompt guidance, 1.5–10. Higher values follow the prompt more closely. Sending this to flux-2-pro or flux-2-max returns a 400.

fallback — object, optional

reAPI may serve a request on either of two upstreams and automatically retries on the second one if the first fails to deliver. Send {"enabled": false} to disable that retry and fail fast instead:

{
  "model": "flux-2-pro",
  "prompt": "A cinematic city at night",
  "fallback": { "enabled": false }
}

Your price is unaffected either way — both upstreams bill the same rate for the same request. Turning the retry off trades availability for a faster failure.

Some requests are served by a single upstream regardless, because only one of the two can honour them exactly and reAPI never relaxes a parameter to make a request routable: steps, guidance, seed, width / height, a pixel-string size, a 3MP / 4MP resolution, prompt_upsampling: true, an explicit output_format, or a safety_tolerance above 2.

Pricing

FLUX.2 bills two terms per request:

credits = ceil((output_usd(tier, resolution) + n_refs × reference_usd(tier)) × 1000)

where 1 credit = $0.001 USD. In words:

1. Output — priced by tier (pro / max / flex) × resolution (1MP … 4MP). Aspect ratio does not change it.
2. Reference images — each distinct image in image_urls adds a per-image charge at the same tier. A request with three references costs meaningfully more than the same prompt with none. Repeating the same URL does not add a charge and does not add a second reference: duplicates collapse before the request is sent, so [a, a, b] is billed and generated as two references.

Each request produces one image. Failed and rejected requests are not charged.

Exact per-tier credit costs surface on the model page and through the playground estimator before submit.

Response

{
  "id": "task_019dfd44b7fd74168541552a3260a623",
  "model": "flux-2-pro",
  "status": "completed",
  "output": {
    "image_urls": [
      "https://cdn.reapi.ai/...jpg"
    ]
  }
}

Generated URLs expire — mirror them to your own storage if you need long-term retention.

Errors

Failures return the standard reAPI envelope { error: { code, message, request_id } }. Common cases:

• Invalid input (empty prompt, more than 8 image_urls, a non-http(s) URL, width without height, output over 4 MP, safety_tolerance outside 0–5, or steps / guidance on a non-Flex tier) → 400.
• Insufficient credits → 402.
• Rate limited → 429.

See the full catalog at /docs/api/errors.

Tips

• Reference images are billed per image. If you are iterating on composition, drop to one reference while you explore and add the rest for the final pass.
• Batch jobs that do not need detail should run at 1MP — the resolution tier is the single biggest lever on cost.
• Use width + height when a layout has fixed dimensions; it saves a downstream crop and guarantees the aspect the design expects.
• Put text you want rendered in the image in quotes inside the prompt.
• On the Flex tier, raise guidance when the render drifts from the prompt and raise steps when it looks under-detailed. Change one at a time.

Related

• FLUX Kontext
• FLUX.2 (alternate surface)
• Tasks API
• Error codes