Wan API (Alibaba) with ruble billing: text-to-video and image-to-video through one Zerocoder key. Per-clip pricing, code samples, limits, refunds.
Wan is Alibaba's video generation family, available through Zerocoder as wan/2-7-text-to-video and wan/2-7-image-to-video. The family turns a text prompt or a single reference frame into a short video clip, which makes it a fit for product demos, social clips and quick previsualization.
Text-to-video builds a scene from a written description end to end, while image-to-video keeps the look of a reference frame and animates it, so you can bring an existing photo or render to life instead of generating a scene from scratch.
One API key and one ruble wallet cover both Wan models and every other model on the platform. You top up the wallet with a Russian card or a company invoice directly in the cabinet, no VPN or foreign card needed. Docs, the key cabinet and the price list live on zerocoder.com, and the HTTP interface is compatible with common SDK and HTTP client setups, so you can call it the same way you'd call any REST API.
You send a request to /v1/videos with the model id, a prompt, the clip length in seconds and, for image-to-video, an https URL of the reference frame. The response comes back with a task id and status queued, and the charge is taken at that moment. You then poll /v1/videos with that id until the status turns into completed, where you get the video URL, or failed, where the money is refunded automatically.
Built from the live GET /v1/models price list: prices include our margin, in rubles and in dollars at the internal rate. Anything missing is temporarily not served by the gateway.
| Model | Unit | ₽ | $ | Cost examples | Endpoint |
|---|---|---|---|---|---|
wan/2-7-text-to-videorecommended-68% | per clip | 14.4 ₽ | $0.160 | 1 generation: 14.4 ₽ · $0.160 10 generations: 144 ₽ · $1.60 | /v1/videos |
wan/2-7-image-to-videorecommended-68% | per clip | 14.4 ₽ | $0.160 | 1 generation: 14.4 ₽ · $0.160 10 generations: 144 ₽ · $1.60 | /v1/videos |
Official vendor list prices against Zerocoder prices for the same unit — a second of video, an image or a million tokens. The “-NN%” badge is the price difference for the same unit (for text, a 3 : 1 input : output blend without the per-call minimum; assumptions are below the table). Only models where we are at least 5% cheaper than the official API are listed here; the full family price list is in the table above.
| Model | Official API (USD) | Zerocoder (₽ and $) | Saving |
|---|---|---|---|
wan/2-7-text-to-videoAlibaba · Wan 2.7 text-to-video | $0.50 per 5 s of video ($0.10/s)720P · ≈ 45 ₽ | 14.4 ₽ · $0.160per clip | -68% |
wan/2-7-image-to-videoAlibaba · Wan 2.7 image-to-video | $0.50 per 5 s of video ($0.10/s)720P · ≈ 45 ₽ | 14.4 ₽ · $0.160per clip | -68% |
Pick a model and a volume — the calculator uses the same formulas as API billing.
Prices from GET /v1/models, internal exchange rate. Failed generations are not charged.
One zc-sk-… key, the x-api-key header (or Authorization: Bearer), base URL https://zerocoder.com/api/v1. Samples are generated from the real route contracts.
Account → API: top up the balance in rubles and create a zc-sk-… key
Copy the sample below — the model id is already taken from the price list.
GET /v1/balance shows the balance; the call price comes back in the response.
# 1) start the render — price_rub is charged up front, refunded if the render fails
curl https://zerocoder.com/api/v1/videos \
-H "x-api-key: zc-sk-…" -H "content-type: application/json" \
-d '{"model": "wan/2-7-text-to-video", "prompt": "A drone shot over a foggy pine forest at sunrise, slow dolly in", "seconds": 5, "aspect": "16:9"}'
# → {"id":"vid_…","object":"video","model":"wan/2-7-text-to-video","status":"queued","seconds":5,"price_rub":…,"balance_rub":…,"poll":"/api/v1/videos?id=vid_…"}
# 2) poll until status is "completed" (url) or "failed" (refunded: true)
curl "https://zerocoder.com/api/v1/videos?id=vid_…" -H "x-api-key: zc-sk-…"
# image-to-video: add "image": "https://…/frame.jpg" to the bodyimport time, requests
API = "https://zerocoder.com/api/v1"
H = {"x-api-key": "zc-sk-…"}
job = requests.post(f"{API}/videos", headers=H, json={
"model": "wan/2-7-text-to-video",
"prompt": "A drone shot over a foggy pine forest at sunrise, slow dolly in",
"seconds": 5, # 4–10
"aspect": "16:9", # or 9:16 / 1:1; "image": "https://…" for image-to-video
}).json()
print("charged", job["price_rub"], "RUB, balance", job["balance_rub"])
while True:
st = requests.get(f"{API}/videos", headers=H, params={"id": job["id"]}).json()
if st["status"] == "completed":
print(st["url"]); break
if st["status"] == "failed":
print("failed, refunded:", st.get("refunded")); break
time.sleep(4)const API = "https://zerocoder.com/api/v1";
const H = { "x-api-key": "zc-sk-…", "content-type": "application/json" };
const job = await fetch(`${API}/videos`, {
method: "POST", headers: H,
body: JSON.stringify({ model: "wan/2-7-text-to-video", prompt: "A drone shot over a foggy pine forest at sunrise, slow dolly in", seconds: 5, aspect: "16:9" }),
}).then((r) => r.json());
console.log("charged", job.price_rub, "RUB");
for (;;) {
const st = await fetch(`${API}/videos?id=${job.id}`, { headers: H }).then((r) => r.json());
if (st.status === "completed") { console.log(st.url); break; }
if (st.status === "failed") { console.log("failed, refunded:", st.refunded); break; }
await new Promise((r) => setTimeout(r, 4000));
}
From a key to production — six steps based on the real /api/v1 contracts. Each step links to the relevant docs section.
Sign up, open the API section of your account, top up the balance in rubles and create a zc-sk-… key. One key unlocks Wan and every other model in the catalogue.
Read more →POST /v1/videos with model, prompt, seconds and aspect: the response carries the job id, status queued and price_rub — the amount is charged before the Wan render starts.
Read more →GET /v1/videos?id=… every few seconds until status is completed (url field) or failed (refunded: true — the money is already back on the balance). The file URL is temporary — download right away.
Read more →30 requests per minute per key: on 429 wait and retry with a growing pause; 402 — top up the balance; 400 — check model against GET /v1/models. Failed generations are not charged, so retrying is safe.
Read more →GET /v1/balance returns the balance and 30-day spend; every call's price comes back in the response — write it to your logs and set an alert threshold in your own monitoring.
Read more →Keep the key in server secrets, never in the frontend. For high Wan volumes contact us — the limit is raised individually and companies can pay by invoice. Check the observed stability on the status page.
Read more →HTTP Request node: POST to the /api/v1 endpoint, x-api-key header and a JSON body built from workflow fields. For video add a Wait node and a second HTTP Request on GET ?id= in a loop until completed.
HTTP “Make a request” module: POST, Body type JSON, x-api-key header. The response is parsed automatically — the image url or the answer text flows into the next modules.
Webhooks by Zapier → Custom Request action: POST, Data — JSON with model and prompt, Headers — x-api-key. Response fields are available to the following Zap steps.
In Apps Script use UrlFetchApp.fetch with method post, contentType application/json and a payload from the cells: prompts in one column, results written to the next one.
A bot on aiogram, grammY or Telegraf calls the same HTTP API: user message → model request → reply in the chat. For video — poll the status and send the file from the finished URL.
For text models the official OpenAI and Anthropic SDKs work with a swapped base URL and our key — your application code stays the same. Images and video are a plain HTTP call from any language.
More than a key: we help wire Wan and other models into your workflows — from audit to production, with invoice payment.
# 401 {"type":"error","error":{"type":"authentication_error","message":"Invalid or revoked API key."}}
# 402 {"type":"error","error":{"type":"billing_error","message":"Not enough API balance: …"}}
# 429 {"type":"error","error":{"type":"rate_limit_error","message":"…"}} ← 30 requests/min per key
# 400 {"type":"error","error":{"type":"invalid_request_error","message":"Unknown model \"…\". See GET /v1/models."}}
# 502 {"type":"error","error":{"type":"api_error","message":"… Charge refunded."}} ← start failed → money back
# GET ?id → {"id":"vid_…","status":"failed","refunded":true} ← render failed → money back
Numbers come from the public /status summary: the share of successful requests over a period for the whole API and for the category as a whole (the summary has no per-model breakdown), no SLA promises. Failed generations are not charged, so retrying is safe.
Tracking since Aug 6, 2026.
An e-commerce team animates static product photos into short turntable-style clips for listings
A social media bot turns daily text prompts into short clips for feed and story posting
A marketing agency builds a text-to-video pipeline that drafts ad concepts from campaign briefs
An indie hacker adds a 'bring your image to life' feature to a photo-editing app
A storyboard tool for video teams previews scenes from script lines before full production
An internal design team generates quick motion mockups from brand stills for review decks
Copy the prompt and the parameters into the request body — the fields are named exactly as in the /api/v1 contract. Replace the angle-bracket placeholders with your material.
Slow aerial glide over terraced rice fields at sunrise, low mist between the hills, soft golden light, calm cinematic mood
seconds: 10aspect: 16:9A long shot without people is a good candidate for 10 seconds.
Close-up of fresh raspberries falling into a glass of sparkling water in slow motion, bubbles rising, bright kitchen light, vertical framing
seconds: 5aspect: 9:16The camera slowly pushes in, curtains move gently in the breeze, sunlight flickers across the room, keep the furniture and layout unchanged
seconds: 5aspect: 16:9image: https://…/room.jpgPick the image-to-video version in the table; the frame link must be a public https URL.
A white sneaker rotates slowly on a concrete pedestal, dramatic side light, subtle dust in the air, premium commercial look
seconds: 5aspect: 1:1Rain drops running down a window at night with blurred city lights behind, slow focus pull, moody blue and amber tones
seconds: 8aspect: 16:9
Industries where this kind of generation becomes part of the workflow rather than an experiment.
Same key, same balance, same request contracts.
Wan is sold per clip: the price of one video for each version is in the table above and in GET /v1/models, in rubles and including our margin. The same amount comes back in the price_rub field of the POST /v1/videos response before rendering starts, so a series is budgeted up front: the clip price times the number of clips.
With requests: POST /v1/videos with the x-api-key header and a JSON body of model, prompt, seconds and aspect — the response returns a job id and status queued. Then poll GET /v1/videos with that id every few seconds until completed and download the mp4 right away. Full Wan samples for Python, Node.js and the shell are in the “First request” block.
They are two separate Wan models with different ids in the table. Text-to-video builds the whole scene from the description; image-to-video takes a start frame from the image field (a public https URL) and animates it while keeping the composition — then describe only the motion in the prompt. Send the picture to the image-to-video version: the text model does not use a start frame.
Duration is set in seconds from 4 to 10 (default 5), the format with aspect 16:9, 9:16, 1:1, 4:3 or 3:4, or an explicit size. The start frame for image-to-video must be a public https URL. The shared key limit is 30 requests per minute across all endpoints; above it Wan returns 429 — retry after a pause.
The Wan clip price is refunded to the API balance automatically: on status failed right away, with refunded: true in the response, and a job that never delivers a result is closed with a refund by the system within a few hours. Only clips that end with status completed and a file link are paid for.
Yes: in the API account a company requests an invoice, pays it by bank transfer, and the amount is credited to the API balance — Wan and the rest of the catalogue work with the same key. A Russian bank card works for an instant top-up; no Alibaba Cloud account or foreign card is needed, and accounting documents are issued on request.
There are no free clips in the API: the wallet starts with a minimum first top-up, and every Wan clip is charged at the table price. Tune the prompt on one or two clips before launching a series — the request format is the same for text-to-video and image-to-video, only model and the image field change.
No: Wan is billed per clip, so a 10-second video costs the same as a 4-second one — duration is set with the seconds field. Keep in mind that a longer clip makes it harder for the model to hold a character and a storyline: start scenes with people at 5 seconds and use longer clips for landscapes, backgrounds and smooth camera glides.
Up to -68% vs the official API · ruble billing, no VPN
