Wan 2.7

Wan 2.7's official protocol is Alibaba Bailian's DashScope asynchronous video synthesis API. The ModelGo gateway accepts the official paths and request/response shapes unchanged: create → poll → fetch content. Which supplier renders the clip is a routing decision inside the gateway and invisible to the caller; when a request lands on the official Bailian channel, your request body is forwarded as you wrote it and the gateway only replaces model (legacy-spelling fields are rewritten to the official spelling).

With the official DashScope SDK, point base_url (DASHSCOPE_HTTP_BASE_URL) at https://api.modelgo.ai and use your ModelGo API key as api_key.

1. Create a task

POST https://api.modelgo.ai/api/v1/services/aigc/video-generation/video-synthesis

Headers: Authorization: Bearer $MODELGO_API_KEY, X-DashScope-Async: enable, Content-Type: application/json.

Text to video

curl https://api.modelgo.ai/api/v1/services/aigc/video-generation/video-synthesis \
  -H "Authorization: Bearer $MODELGO_API_KEY" \
  -H "X-DashScope-Async: enable" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.7-t2v",
    "input": {"prompt": "A fluffy orange kitten playing with a ball of yarn on a sunny windowsill"},
    "parameters": {"resolution": "1080P", "duration": 5, "prompt_extend": true}
  }'

Image to video

curl https://api.modelgo.ai/api/v1/services/aigc/video-generation/video-synthesis \
  -H "Authorization: Bearer $MODELGO_API_KEY" \
  -H "X-DashScope-Async: enable" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.7-i2v",
    "input": {
      "prompt": "The cat slowly tilts its head and blinks",
      "media": [{"type": "first_frame", "url": "https://example-cdn.com/cat.jpg"}]
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

Response

{
  "output": {"task_id": "gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5", "task_status": "PENDING", "submit_time": "2026-09-07 17:05:59.000"},
  "request_id": "gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5"
}

output.task_id is the gateway task id; use it to poll and to fetch content.

2. Parameters

parameters keys not listed above (for example shot_type, deprecated on wan2.7) are not interpreted by the gateway and do not cause an error: as in the official API the upstream ignores them, and billing is unaffected.

Audio

• The official wan2.7 API has no audio switch, and ModelGo follows it: without a driving_audio in input.media[] (input.audio_url on t2v) the model generates matching background music or sound effects from the picture; with one, your audio drives the soundtrack.
• So wan2.7 has a single price (the audio tier) — there is no "silent tier". parameters.audio is not a wan2.7 parameter and the gateway does not send it upstream.
• A request that explicitly asks for silence is refused: parameters.audio=false or a top-level generate_audio=false / "false" returns 400 at create time, costs nothing, and is never turned into a clip with sound that is then billed. true is the same as omitting it and causes no error.

Resolution tiers

• For 4:3 / 3:4, size also accepts Bailian's spelling: 720P 1104*832 / 832*1104, 1080P 1648*1248 / 1248*1648 — the same tier and price as the spelling in the table.
• Only 720P / 1080P are offered. 4K or a size outside the table is refused at create time (HTTP 400); nothing is rendered at a lower tier silently and nothing is charged.
• Routed by deliverable tier. The gateway only hands a request to a supplier that renders the stated tier (for example wan2.7-t2v at 720P is currently rendered by the official Bailian lane only); no supplier renders 1080P in its place while 720P is billed. If no supplier can render the tier right now the create returns 400 and nothing is charged.
• input keys not in the table above (e.g. ref_images_url) are not supported on this endpoint and are refused rather than ignored.

Video continuation (first_clip, i2v only)

• Send {"type": "first_clip", "url": "<video URL>"} in input.media[] (or the flat input.first_clip_url) to continue from an existing video. The clip must be a 2–10 s mp4 / mov at a publicly downloadable URL (not an authenticated address such as this gateway's own /v1/videos/{id}/content).
• wan2.7-i2v only. It cannot be combined with first_frame (img_url) or driving_audio (audio_url); it may be paired with last_frame. Any violation is refused at creation with 400 and costs nothing.
• Billed as i2v, at the same tier price as a non-continuation i2v call.
• Not every supplier channel can carry a continuation: the gateway only routes it to channels confirmed to support it, and returns 400 (no charge) when none is available.

First-frame image requirements (i2v)

• A direct link the upstream can GET: no auth, no redirects, no Referer / Cookie dependency.
• Some hosts return 403 to non-browser fetches; the task then fails (task_status=FAILED) and is not billed. Use your own object storage or CDN.

3. Poll the task

GET https://api.modelgo.ai/api/v1/tasks/{task_id}

curl https://api.modelgo.ai/api/v1/tasks/gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5 \
  -H "Authorization: Bearer $MODELGO_API_KEY"

When finished:

{
  "output": {
    "task_id": "gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5",
    "task_status": "SUCCEEDED",
    "submit_time": "2026-09-07 17:05:59.000",
    "end_time": "2026-09-07 17:09:06.000",
    "video_url": "https://api.modelgo.ai/v1/videos/gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5/content"
  },
  "usage": {"video_duration": 6, "video_count": 1},
  "request_id": "gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5"
}

• task_status: PENDING → RUNNING → SUCCEEDED / FAILED. On failure output.code / output.message carry the reason and the task is not billed.
• Generation usually takes 1–3 minutes; poll every 5–10 s with backoff.
• output.video_url points at the gateway content endpoint and requires the API key. Do not persist it; artifacts have a retention period.

4. Download the video

GET https://api.modelgo.ai/v1/videos/{task_id}/content

curl https://api.modelgo.ai/v1/videos/gw_e99647a0-692d-4fdd-ae46-e1d777e72bf5/content \
  -H "Authorization: Bearer $MODELGO_API_KEY" -o cat.mp4

Streams video/mp4, Range supported. Only the API key that owns the task (or a console session of the same tenant) may fetch it; 409 while not completed, 404 once expired or unknown.

5. Full example (Python, SDK style)

import os, time, requests

BASE = "https://api.modelgo.ai"
H = {"Authorization": f"Bearer {os.environ['MODELGO_API_KEY']}", "X-DashScope-Async": "enable"}

def synthesize(model: str, inp: dict, params: dict, out: str, timeout: int = 600) -> None:
    r = requests.post(f"{BASE}/api/v1/services/aigc/video-generation/video-synthesis",
                      json={"model": model, "input": inp, "parameters": params}, headers=H, timeout=30)
    r.raise_for_status()
    task_id = r.json()["output"]["task_id"]
    deadline = time.time() + timeout
    while True:
        if time.time() > deadline:
            raise TimeoutError(task_id)
        o = requests.get(f"{BASE}/api/v1/tasks/{task_id}", headers=H, timeout=30).json()["output"]
        if o["task_status"] == "SUCCEEDED":
            break
        if o["task_status"] == "FAILED":
            raise RuntimeError(o.get("message") or o.get("code"))
        time.sleep(8)
    with requests.get(o["video_url"], headers=H, stream=True, timeout=120) as c:
        c.raise_for_status()
        with open(out, "wb") as f:
            for chunk in c.iter_content(1 << 20):
                f.write(chunk)

synthesize("wan2.7-t2v", {"prompt": "A fluffy orange kitten playing with yarn"}, {"resolution": "1080P", "duration": 5}, "t2v.mp4")
synthesize("wan2.7-i2v", {"prompt": "The cat slowly tilts its head and blinks", "media": [{"type": "first_frame", "url": "https://example-cdn.com/cat.jpg"}]}, {"resolution": "720P", "duration": 5}, "i2v.mp4")

6. Billing

• Video seconds × resolution-tier unit price. The requested duration is pre-authorized at create time; the task settles on the delivered seconds (usage.video_duration). The upstream may deliver one extra second (5 s requested, 6 s delivered) — 6 s are billed.
• A request without a duration is pre-authorized at the default (Wan / Kling 5 s, Veo 8 s) and that default is sent to the supplier; settlement uses the supplier-reported delivered length rounded half-up to whole seconds (5.041 → 5, 5.5 → 6), which may differ from the requested duration by ±1 s.
• A refused create (400) or a failed generation (FAILED) is not billed and the hold is released.
• wan2.7 has no silent tier: every request is billed at the audio-tier unit price (see "Audio" above).
• Unit prices are those shown in the console model list.

Common errors