per generated second for 5–20 s HD or FHD clips — one OpenAI-compatible endpoint at
https://api.apimart.ai/v1, no monthly plan required. (observed 2026-10-12)
Get an API key · Live pricing · Model page · ⚡ 60-second quickstart
Why teams call FLUX 3 Video (flux-3-video) through APIMart
- One key, entire catalog. The same
https://api.apimart.ai/v1base URL andAuthorizationheader reach FLUX 3 Video (flux-3-video) and 300+ other image, video and language models — switch themodelfield, not your client. - Per-second billing. A job is charged for the seconds it actually generates at the resolution you picked.
- The charge comes back in the response. Every completed task reports the amount billed (
cost/credits_cost); a task markedfailedis refunded. - Async by design. Submit, take the
task_id, pollGET /v1/tasks/{id}— batching and retries are ordinary queue work, not a bespoke integration.
FLUX 3 Video covers text-to-video, ordered keyframes, video continuation and synced audio in 5–20 s HD or FHD clips.
| Route | model value |
Billing | Notes |
|---|---|---|---|
| Text-to-video / image-to-video | flux-3-video |
per generated second, by resolution | see the model page for the exact tiers |
Endpoint: POST https://api.apimart.ai/v1/videos/generations (OpenAI-compatible), then poll GET /v1/tasks/{task_id}.
| Resolution | List price | Effective price |
|---|---|---|
| per generated second | — | see the live pricing page |
Prices are a snapshot; the pricing page and data/model.json are refreshed by CI, and a completed task reports the exact amount in its cost field.
| Field | Type | Default | Notes |
|---|---|---|---|
model |
string | required | the model id shown in the route table above |
prompt |
string | required | the shot description; camera and subject motion belong here |
duration |
integer | route default | clip length in seconds — billing is per generated second |
resolution |
string | route default | 480P / 720P / 1080P or the route’s own tiers |
size / aspect_ratio |
string | 16:9 |
output framing |
image_urls |
string[] | — | first frame, first + last frame, or style references, where supported |
seed |
integer | — | fix the seed to reproduce a take |
The table lists the fields accepted by the APIMart OpenAI-compatible video endpoint. Check the model page for anything this model adds on top.
curl --request POST --url https://api.apimart.ai/v1/videos/generations \
--header "Authorization: Bearer $APIMART_API_KEY" --header 'Content-Type: application/json' \
-d '{"model":"flux-3-video","prompt":"A bamboo forest at sunrise, slow push-in","duration":5,"resolution":"720P","size":"16:9"}'import os, time, requests
BASE = "https://api.apimart.ai/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}", "Content-Type": "application/json"}
created = requests.post(f"{BASE}/videos/generations", headers=HEADERS, timeout=60, json={
"model": "flux-3-video", "prompt": "A bamboo forest at sunrise, slow push-in",
}).json()
task_id = created["data"]["id"]
while True:
task = requests.get(f"{BASE}/tasks/{task_id}", headers=HEADERS, timeout=60).json()["data"]
if task["status"] in ("completed", "failed"):
break
time.sleep(5)
print(task.get("cost"), task["result"]["videos"][0]["url"])Runnable versions: examples/. The task lifecycle is pending → processing → completed | failed.
What is the FLUX 3 Video model id?
flux-3-video. See the route table above.
How is it billed?
per generated second for 5–20 s HD or FHD clips. The completed task reports the exact amount in its cost field; a failed task is refunded.
How do I get a key?
Use the Get an API key link — $1 minimum top-up, no plan required.
| Purpose | Attributed link | Target |
|---|---|---|
| Model page | https://go.apimart.ai/k-b63e6d | apimart.ai/zh/model/flux-3-video |
| Live pricing | https://go.apimart.ai/k-850221 | apimart.ai/zh/pricing |
| Get an API key | https://go.apimart.ai/k-7c040e | apimart.ai/zh/keys |
Outbound APIMart links are minted through the promo link API; hand-made tracking parameters are rejected by
tools/check_links.py in CI.
FLUX 3 Video is a third-party model served through APIMart; this repository documents how to call it and publishes model ids and prices, and does not claim official status. Model names, prices and documentation belong to their respective owners.
MIT — see LICENSE.