Video Model Guide

Video generation is asynchronous. Request body shape differs by supplier — use the matching protocol tab.

1. Basic Calling Information

Item	Description
Base URL	`https://tokenhub.link`
Endpoints	`POST /v1/video/generations` (create task) `GET /v1/video/tasks/{task_id}?model=<model>` (poll)
Authentication	`Authorization: Bearer <TokenHub API Key>`. Create the key in API Keys in console.
Model Identifier	Use `model` in create body. See Model Catalog — e.g. `wan2.7-t2v`, `tokenhub/happyhorse-1.0-t2v`.
Polling	Poll every 10–15s until terminal state. Pass the same `model` on poll URL.

2. Request body by supplier protocol

Create then poll. Pick the tab matching your model supplier.

2.1 Create Task

curl -sS "https://tokenhub.link/v1/video/generations" \
  -H "Authorization: Bearer $TOKENHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tokenhub/happyhorse-1.0-t2v",
    "prompt": "一只猫在草地奔跑，广角镜头，风格写实",
    "duration": 5,
    "width": 1280,
    "height": 720
  }'

You may also pass resolution + ratio (e.g. 720p + 16:9); the gateway converts them to width/height.

Create response:

{
  "id": "task_xxxxxxxx",
  "object": "video.generation.task",
  "status": "pending",
  "model": "tokenhub/happyhorse-1.0-t2v"
}

2.2 Query Task (Polling)

curl -sS "https://tokenhub.link/v1/video/tasks/task_xxxxxxxx?model=tokenhub/happyhorse-1.0-t2v" \
  -H "Authorization: Bearer $TOKENHUB_API_KEY"

2. End-to-End Call and Response Example

2.1 Create Task

curl --location 'https://tokenhub.link/v1/video/generations' \
                      -H 'X-DashScope-Async: enable' \
                      -H "Authorization: Bearer $TOKENHUB_API_KEY" \
                      -H 'Content-Type: application/json' \
                      -d '{
                      "model": "wan2.7-t2v",
                      "input": {
                          "prompt": "A tense detective investigation with cinematic storytelling. Shot 1 [0-3s], wide shot: rainy New York street with neon lights as a detective in a black trench coat walks quickly. Shot 2 [3-6s], medium shot: the detective enters an old building and the door slowly closes behind him. Shot 3 [6-9s], close-up: focused eyes, distant sirens, and a subtle frown. Shot 4 [9-12s], medium shot: he moves carefully through a dim corridor with a flashlight. Shot 5 [12-15s], close-up: he discovers a key clue and shows a sudden expression of realization."
                      },
                      "parameters": {
                          "resolution": "720P",
                          "ratio": "16:9",
                          "prompt_extend": true,
                          "watermark": true,
                          "duration": 15
                      }
                  }'

Create-task response example:

{
                "output": {
                    "task_status": "PENDING",
                    "task_id": "0385dc79-5ff8-4d82-bcb6-xxxxxx"
                },
                "request_id": "4909100c-7b5a-9f92-bfe5-xxxxxx"
            }

2.2 Query Task (Polling)

curl -X GET 'https://tokenhub.link/v1/video/tasks/{task_id}?model=wan2.6-t2v' \
            --header "Authorization: Bearer $TOKENHUB_API_KEY"

Polling response example (success):

{
                      "request_id": "caa62a12-8841-41a6-8af2-xxxxxx",
                      "output": {
                          "task_id": "eff1443c-ccab-4676-aad3-xxxxxx",
                          "task_status": "SUCCEEDED",
                          "submit_time": "2025-09-29 14:18:52.331",
                          "scheduled_time": "2025-09-29 14:18:59.290",
                          "end_time": "2025-09-29 14:23:39.407",
                          "orig_prompt": "An epic yet adorable scene: a small cartoon cat general in detailed golden armor stands bravely on a cliff, riding a tiny but heroic warhorse and reciting a dramatic battle poem. Below, a massive army of mice charges forward with improvised weapons. The scene blends humor, cuteness, and epic war-film atmosphere under dark clouds above distant snowy mountains.",
                          "video_url": "https://dashscope-result-sh.oss-accelerate.aliyuncs.com/xxx.mp4?Expires=xxx"
                      },
                      "usage": {
                          "duration": 10,
                          "input_video_duration": 0,
                          "output_video_duration": 10,
                          "video_count": 1,
                          "ratio": "16:9",
                          "SR": 720
                      }
            }

3. Detailed Parameters

3.1 Top-level Request Body

Field	Type	Required	Description
`model`	`string`	Yes	Model name, such as `wan2.6-t2v`, `wan2.5-t2v-preview`, `wan2.2-t2v-plus`.
`input`	`object`	Yes	Input object. Core field is `prompt`, and some models also support `audio_url`.
`parameters`	`object`	No	Video generation control parameters, see 3.3.

3.2 input Fields

Field	Type	Required	Description
`input.prompt`	`string`	Yes	Video generation prompt. Different models may have different limits on length, language, and style.
`input.negative_prompt`	`string`	No	Negative prompt.
`input.audio_url`	`string(URL)`	Conditional	Some models support audio-driven generation via `audio_url`; do not pass it when unsupported.

3.3 parameters Fields

Field	Type	Required	Description
`size`	`string`	No	Resolution in pixels, such as `1280720` and `832480`.
`duration`	`integer`	No	Video duration in seconds. Supported range varies by model.
`prompt_extend`	`boolean`	No	Whether to enable prompt expansion.
`shot_type`	`string`	No	Shot mode, for example `multi` (supported only by some models).
`watermark`	`boolean`	No	Whether to add watermark (depends on model support).
`seed`	`integer`	No	Random seed for approximate reproducibility.

3.4 Task Status and Polling Recommendations

Common task states: PENDING / RUNNING / SUCCEEDED / FAILED / CANCELED / UNKNOWN.
Recommended polling interval is around 10-15 seconds; query validity for task_id is typically about 24 hours.
After success, download video_url promptly because links are usually temporary.

4. Scenario Examples

4.1 Multi-shot Narrative

curl --location 'https://tokenhub.link/v1/video/generations' \
                      -H 'X-DashScope-Async: enable' \
                      -H "Authorization: Bearer $TOKENHUB_API_KEY" \
                      -H 'Content-Type: application/json' \
                      -d '{
                      "model": "wan2.7-t2v",
                      "input": {
                          "prompt": "A tense detective investigation with cinematic storytelling. Shot 1 [0-3s], wide shot: rainy New York street with neon lights as a detective in a black trench coat walks quickly. Shot 2 [3-6s], medium shot: the detective enters an old building and the door slowly closes behind him. Shot 3 [6-9s], close-up: focused eyes, distant sirens, and a subtle frown. Shot 4 [9-12s], medium shot: he moves carefully through a dim corridor with a flashlight. Shot 5 [12-15s], close-up: he discovers a key clue and shows a sudden expression of realization."
                      },
                      "parameters": {
                          "resolution": "720P",
                          "ratio": "16:9",
                          "prompt_extend": true,
                          "watermark": true,
                          "duration": 15
                      }
            }'

4.2 Auto Voiceover

curl --location 'https://tokenhub.link/v1/video/generations' \
                      -H 'X-DashScope-Async: enable' \
                      -H "Authorization: Bearer $TOKENHUB_API_KEY" \
                      -H 'Content-Type: application/json' \
                      -d '{
                      "model": "wan2.7-t2v",
                      "input": {
                          "prompt": "An epic yet adorable scene: a small cartoon cat general in detailed golden armor stands bravely on a cliff, riding a tiny but heroic warhorse and reciting a dramatic battle poem. Below, a massive army of mice charges forward with improvised weapons. The scene blends humor, cuteness, and epic war-film atmosphere under dark clouds above distant snowy mountains."
                      },
                      "parameters": {
                          "resolution": "720P",
                          "ratio": "16:9",
                          "prompt_extend": true,
                          "duration": 10
                      }
            }'

4.3 Custom Audio Input

curl --location 'https://tokenhub.link/v1/video/generations' \
                      -H 'X-DashScope-Async: enable' \
                      -H "Authorization: Bearer $TOKENHUB_API_KEY" \
                      -H 'Content-Type: application/json' \
                      -d '{
                      "model": "wan2.7-t2v",
                      "input": {
                          "prompt": "An epic yet adorable scene: a small cartoon cat general in detailed golden armor stands bravely on a cliff, riding a tiny but heroic warhorse and reciting a dramatic battle poem. Below, a massive army of mice charges forward with improvised weapons. The scene blends humor, cuteness, and epic war-film atmosphere under dark clouds above distant snowy mountains.",
                          "audio_url": "https://example.com/audio/cinematic-narration.mp3"
                      },
                      "parameters": {
                          "resolution": "720P",
                          "ratio": "16:9",
                          "prompt_extend": true,
                          "duration": 10
                      }
            }'

4.4 Using Negative Prompt

curl --location 'https://tokenhub.link/v1/video/generations' \
                      -H 'X-DashScope-Async: enable' \
                      -H "Authorization: Bearer $TOKENHUB_API_KEY" \
                      -H 'Content-Type: application/json' \
                      -d '{
                      "model": "wan2.7-t2v",
                      "input": {
                          "prompt": "A small cat runs under moonlight.",
                          "negative_prompt": "flowers"
                      },
                      "parameters": {
                          "resolution": "720P",
                          "ratio": "16:9"
                      }
            }'

5. Reference Links

2.1 Create Task

curl -sS "https://tokenhub.link/v1/video/generations" \
  -H "Authorization: Bearer $TOKENHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ark/doubao-seedance-1-0-pro-250528",
    "prompt": "海浪拍打礁石，慢镜头，电影质感",
    "ratio": "16:9",
    "duration": 5
  }'

Alternative: content: [{ "type": "text", "text": "..." }] instead of prompt.

2.2 Query Task (Polling)

curl -sS "https://tokenhub.link/v1/video/tasks/{task_id}?model=ark/doubao-seedance-1-0-pro-250528" \
  -H "Authorization: Bearer $TOKENHUB_API_KEY"

Volcengine Ark model list