De zéro à une image générée en quelques minutes.
Collez-le dans ChatGPT, Claude, Gemini ou votre agent de codage : il contient tout ce qu'il faut pour ajouter l'API à votre projet.
Add the PicassoIA API to my project, to generate images and videos from my own code.
How the API works:
- Base URL: https://api.picassoia.com/v1. JSON requests and responses.
- Auth: send `Authorization: Bearer <key>`, reading the key from the PICASSOIA_API_KEY environment variable. Keep it server-side: the API sends no CORS headers, so never call it from a browser or put the key in an app.
- Create a prediction: `POST /v1/models/{owner}/{name}/predictions` with the body `{"input": {...}}`. It answers `201` with the prediction in the `starting` state and never waits for the result.
- Poll `prediction.urls.get`, waiting `eta.next_poll_in_seconds` between calls, until `status` is `succeeded`, `failed` or `canceled`. Then read `output`, or `error` when it failed.
- `GET /v1/models` lists every model with its input schema (JSON Schema). Unknown fields are refused with `422 invalid_input`.
- Images go as public https URLs or data URLs (PNG, JPEG or WebP, up to 5 MB each); the request body is limited to 10 MB.
- Up to 5 predictions queued or running per account; the next one answers `429 concurrency_limit` with a `Retry-After` header.
- Output URLs are hosted by PicassoIA: download the files you want to keep.
- Docs: https://picassoia.com/en/api
Models:
- `picassoia/seedance-2.5-lite` (Seedance 2.5 Lite): text or image to video with synchronized audio: 5, 10 or 15 s at 480p, up to 10 s at 720p; optional closing frame (last_frame_image, needs image). Output: one MP4 URL (a string).
- `picassoia/picassoia-video` (PicassoIA Video): text or image to video: 480p up to 20 s, 720p up to 10 s, 1080p up to 5 s. Output: one MP4 URL (a string).
- `picassoia/picassoia-image` (PicassoIA Image): text to image: 1 or 2 images from a prompt. Output: a list of image URLs.
- `picassoia/picassoia-image-editor-pro` (PicassoIA Image Editor Pro): edits or combines 1 to 4 images following a prompt (refer to them as "image 1", "image 2"…). Output: a list of image URLs.
- `picassoia/picassoia-reference-to-video` (PicassoIA Reference To Video): reference to video with synchronized audio: up to 9 reference images, an optional reference video and an optional reference audio (a voice a character lip-syncs, or music), named in the prompt as "Image 1", "Video 1", "Audio 1". At least one image or the video is required; the video and the audio go as public https URLs only. 480p or 720p, 1 to 15 s. Output: one MP4 URL (a string).Créez une clé dans votre compte et conservez-la sur votre serveur, dans la variable d'environnement PICASSOIA_API_KEY. La création de clés et de prédictions nécessite un forfait Infinite actif.
export PICASSOIA_API_KEY="pia_sk_…"Créez une prédiction avec un seul POST. L'API répond généralement en quelques secondes, sans attendre le résultat.
curl -s -X POST https://api.picassoia.com/v1/models/picassoia/picassoia-image/predictions \
-H "Authorization: Bearer $PICASSOIA_API_KEY" \
-H "Content-Type: application/json" \
-d '{"input":{"prompt":"a lighthouse at sunset, oil painting"}}'const API = 'https://api.picassoia.com/v1'
const headers = {
Authorization: `Bearer ${process.env.PICASSOIA_API_KEY}`,
'Content-Type': 'application/json',
}
const input = {
prompt: "a lighthouse at sunset, oil painting"
}
const created = await fetch(`${API}/models/picassoia/picassoia-image/predictions`, {
method: 'POST',
headers,
body: JSON.stringify({ input }),
})
let prediction = await created.json()
if (!created.ok) throw new Error(`${prediction.code}: ${prediction.detail}`)
console.log(prediction.id, prediction.status) // api_…, startingimport os, requests
API = "https://api.picassoia.com/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['PICASSOIA_API_KEY']}"}
payload = {"input": {
"prompt": "a lighthouse at sunset, oil painting",
}}
response = requests.post(f"{API}/models/picassoia/picassoia-image/predictions", json=payload, headers=HEADERS)
prediction = response.json()
if not response.ok:
raise RuntimeError(f"{prediction['code']}: {prediction['detail']}")
print(prediction["id"], prediction["status"]) # api_…, starting{
"id": "api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60",
"model": "picassoia/picassoia-image",
"input": {
"prompt": "a lighthouse at sunset, oil painting",
"aspect_ratio": "1:1",
"num_outputs": 1,
"output_format": "jpg",
"output_quality": 80
},
"output": null,
"error": null,
"status": "starting",
"created_at": "2026-10-01T09:30:00.000Z",
"started_at": null,
"completed_at": null,
"metrics": {
"predict_time": null
},
"urls": {
"get": "https://api.picassoia.com/v1/predictions/api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60",
"cancel": "https://api.picassoia.com/v1/predictions/api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60/cancel"
},
"eta": {
"seconds": 12,
"next_poll_in_seconds": 6
}
}Interrogez l'URL urls.get de la prédiction, en attendant eta.next_poll_in_seconds entre deux appels, jusqu'à ce que son état soit succeeded, failed ou canceled. Si elle a réussi, téléchargez la sortie.
# Repeat after eta.next_poll_in_seconds until status is
# succeeded, failed or canceled.
curl -s https://api.picassoia.com/v1/predictions/api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60 \
-H "Authorization: Bearer $PICASSOIA_API_KEY"
# The last poll holds the result: "output" when "status" is succeeded,
# "error" when it is failed. With jq, just those three fields:
curl -s https://api.picassoia.com/v1/predictions/api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60 \
-H "Authorization: Bearer $PICASSOIA_API_KEY" | jq '.status, .output, .error'const sleep = (s) => new Promise((resolve) => setTimeout(resolve, s * 1000))
while (!['succeeded', 'failed', 'canceled'].includes(prediction.status)) {
await sleep(prediction.eta?.next_poll_in_seconds ?? 2)
prediction = await (await fetch(prediction.urls.get, { headers })).json()
}
if (prediction.status !== 'succeeded') throw new Error(prediction.error ?? prediction.status)
console.log(prediction.output)import time
while prediction["status"] not in ("succeeded", "failed", "canceled"):
time.sleep((prediction.get("eta") or {}).get("next_poll_in_seconds", 2))
prediction = requests.get(prediction["urls"]["get"], headers=HEADERS).json()
if prediction["status"] != "succeeded":
raise RuntimeError(prediction["error"] or prediction["status"])
print(prediction["output"]){
"id": "api_3f9c2b7e8d1a4c6f9e0b1a2c3d4e5f60",
"status": "succeeded",
"output": [
"https://…/output-1.jpg"
],
"error": null,
"completed_at": "2026-10-01T09:30:08.000Z",
"metrics": {
"predict_time": 6.4
},
"eta": null
}Chaque modèle avec son playground, son code et ses paramètres
Texte et image vers vidéo avec audio, des clips de 5, 10 ou 15 secondes et une sortie en 480p ou 720p.
Texte et image vers vidéo avec audio, des clips de 1 à 20 secondes et une sortie en 480p, 720p ou 1080p.
Texte vers image avec 1 ou 2 images par requête et 7 rapports d'aspect.
Édition d'images par IA avec jusqu'à 4 images d'entrée, 1 ou 2 images par requête et 7 rapports d'aspect.