How to Use FIBO 1.5 API: Generate and Edit in 5 Minutes

TL;DR

• FIBO 1.5 is a generate-plus-edit pair: fibo-gen-1.5/text-to-image for photoreal generation (with accurate in-image text) and fibo-edit-1.5/image-to-image for one-line editing (up to 4 reference images, plus masked local edits). Try it on the Gen model page; new accounts get 200 free credits.
• Both cost the same: $0.0575 per image — a generate-then-edit pipeline is $0.115 (see the pricing page).
• Everything runs on HiAPI /v1/tasks: submit for a taskId, then poll. The editing field is called instruction, not prompt.

What it does

• Photoreal output with accurate text: product labels and storefront signs come out letter-perfect (the label below reads NOIR 07).
• One-line edits: describe the change in instruction; anything you don't mention stays as-is.
• Up to 4 images per composite: try-ons, background swaps, product-into-scene — image_urls are referenced in order.
• Masked local edits: white marks what changes, black stays.

Here's a real Gen output:

Luxury product photo of an amber glass perfume bottle on wet black marble, gold cap, soft rim light, droplets on the glass, the label reads 'NOIR 07' in elegant serif letters

fibo-gen-1.5/text-to-image, 4:5, 768×960, $0.0575.

Try this prompt →

Two endpoints: which to use

Typical pipeline: generate the base in Gen, then produce the set in Edit (new backgrounds, recolors, added elements). Scene templates live in the prompt guide.

Two ways to start

Browser, 3 steps: open the model page → paste the prompt, pick a ratio → generate. One image is $0.0575.

API, 3 steps: POST /v1/tasks to submit → take the taskId → GET /v1/tasks/<taskId> until success.

curl: your first image

curl -X POST https://api.hiapi.ai/v1/tasks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: your-unique-key-001" \
  -d '{
  "model": "fibo-gen-1.5/text-to-image",
  "input": {
    "prompt": "Luxury product photo of an amber glass perfume bottle on wet black marble, gold cap, soft rim light, droplets on the glass, the label reads 'NOIR 07' in elegant serif letters",
    "aspect_ratio": "4:5",
    "seed": 2026
  }
}'

For editing, switch model to fibo-edit-1.5/image-to-image and put the change in instruction, referencing images in order (a two-image try-on):

curl -X POST https://api.hiapi.ai/v1/tasks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "fibo-edit-1.5/image-to-image",
  "input": {
    "image_urls": [
      "https://static.hiapi.ai/gallery/2026/10/78f963cb21d0242c.jpg",
      "https://static.hiapi.ai/gallery/2026/10/36246b7cabde4a1a.jpg"
    ],
    "instruction": "Dress the woman in image 1 with the mustard corduroy jacket from image 2, worn open over her white t-shirt. Keep her pose, face and the studio background.",
    "aspect_ratio": "3:4"
  }
}'

Masked local edit: mask_url pairs with exactly one image, and white marks what changes (the opposite of Ideogram 4.5 Edit):

curl -X POST https://api.hiapi.ai/v1/tasks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "fibo-edit-1.5/image-to-image",
  "input": {
    "image_urls": ["https://static.hiapi.ai/gallery/2026/10/d9037da1559e78a6.jpg"],
    "mask_url": "https://static.hiapi.ai/gallery/2026/10/cae371f9e4689db5.png",
    "instruction": "Replace the floor under the sneaker with light wooden planks."
  }
}'

Python: submit, poll, download

import time, uuid, requests

BASE = "https://api.hiapi.ai/v1/tasks"
HEADERS = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
}

payload = {
    "model": "fibo-gen-1.5/text-to-image",
    "input": {
        "prompt": "Luxury product photo of an amber glass perfume bottle on wet black marble, gold cap, soft rim light, droplets on the glass, the label reads 'NOIR 07' in elegant serif letters",
        "aspect_ratio": "4:5",
        "seed": 2026,
    },
}

resp = requests.post(
    BASE,
    headers={**HEADERS, "Idempotency-Key": str(uuid.uuid4())},
    json=payload,
    timeout=30,
)
resp.raise_for_status()
task_id = resp.json()["taskId"]

while True:
    task = requests.get(f"{BASE}/{task_id}", headers=HEADERS, timeout=30).json()
    print(task["status"])
    if task["status"] in ("success", "fail"):
        break
    time.sleep(3)

if task["status"] == "success":
    image_url = task["output"][0]["url"]   # temporary URL — download promptly
    img = requests.get(image_url, timeout=60)
    with open("output_image", "wb") as f:
        f.write(img.content)
    print("saved")

Parameters

fibo-gen-1.5/text-to-image

fibo-edit-1.5/image-to-image

Any field outside the tables returns a 400.

Errors you'll actually hit (verbatim)

Re-submitting the same Idempotency-Key returns 409 or 422; use a fresh key.

Before you ship

• Callbacks: pass "callback": {"url": "https://..."} at the request top level.
• Persist outputs: output[0].url is temporary — download it right away.
• Name images in instructions: reference image 1 / image 2 in the order of image_urls.
• Mask polarity: FIBO is white-edits, black-keeps — invert your masks when porting scripts from other editors.

FAQ

• Yes, the editing field is called instruction, not prompt; the wrong name returns a missing-field error.
• Yes, four input images max — the fifth returns a 400.
• Yes, seed defaults to 5555 and is fixed — identical inputs reproduce; pass a new seed for a different take.
• Yes, both endpoints share one price and calling convention — change only model and input.

Next: take the perfume prompt and generate a base image, then reshape it with one instruction; scene templates are in the prompt guide.