Skip to content

Vega Video 1.0 Pro โ€‹

Updated

โ† Back to the Vega Video overview for version comparison and selection.

Endpoint: POST https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-videos/tasksAuthentication: Tokenhub API Key / Bearer

This page covers all three variants of the 1.0 Pro series: standard wand-vega-video-1.0-pro, fast wand-vega-video-1.0-pro-fast, and lightweight wand-vega-video-1.0-pro-mini. Their input limits, request parameters, and response shapes are identical โ€” only the model value differs. They differ in generation speed and cost, so pick what fits.

Allowlist registration required

The Pro series can only be called after allowlisting. Contact the WAND team to be registered.

The three variants โ€‹

Variantmodel valuePositioning
Standardwand-vega-video-1.0-proQuality first; the default choice
Fastwand-vega-video-1.0-pro-fastFaster output, suited to batches and previews
Miniwand-vega-video-1.0-pro-miniLightweight tier, lower cost

Basic information โ€‹

ItemValue
model valueswand-vega-video-1.0-pro / wand-vega-video-1.0-pro-fast / wand-vega-video-1.0-pro-mini
Resolution480P / 720P / 1080P / 2K / 4K, default 720P
Ratio21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive, default adaptive
DurationAn integer from 4 to 15 seconds, or -1 (model decides), default 5
CapabilitiesText-to-video / first-and-last-frame / reference-based
Material referenceSupports asset://<id>

Input modes โ€‹

Modecontent compositionCondition
Text-to-videotype=textAt least 1 non-empty text element
First-and-last frametext + image_urlAt most 1 first_frame and 1 last_frame; text is required
Reference-basedtext + image / video / audioEvery media element must declare the matching role
Registered material referencetext + asset://<id>Material status must be DONE, and under the same AppID as the video request

Request parameters โ€‹

ParameterRequiredTypeDescription
modelYesStringwand-vega-video-1.0-pro / wand-vega-video-1.0-pro-fast / wand-vega-video-1.0-pro-mini
contentYesArray[Object]Multimodal input array; must contain at least 1 type=text element
resolutionNoString480P / 720P / 1080P / 2K / 4K. Default 720P
ratioNoString21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive. Default adaptive
durationNoIntegerAn integer from 4 to 15, or -1. Default 5
optionsNoObjectOption bag. Can be omitted entirely; omitting it or passing {} applies all defaults
options.return_last_frameNoBooleanWhether to return the last-frame image. Default false. content.last_frame_url appears only when the model actually produced a last frame
options.return_vega_usageNoBooleanWhether the query endpoint returns usage statistics. Default false. When true, a successful response carries a usage block

content elements โ€‹

content is a union-typed array keyed by type, and the element fields are mutually exclusive. Within one element use only the content field matching its type โ€” never mix different media fields in the same element.

ParameterRequiredTypeDescription
typeYesStringtext / image_url / video_url / audio_url
textConditionalStringRequired and non-empty when type=text. Multiple text elements are concatenated in array order
image_urlConditionalObjectRequired when type=image_url, shaped {"url": "..."}
image_url.urlConditionalStringImage address. Supports a public HTTP/HTTPS URL, inline Base64 (data:<mediatype>;base64,<payload>), and asset://<id>
video_urlConditionalObjectRequired when type=video_url, shaped {"url": "..."}
video_url.urlConditionalStringVideo address. Public HTTP/HTTPS URL only โ€” inline Base64 is not supported
audio_urlConditionalObjectRequired when type=audio_url, shaped {"url": "..."}
audio_url.urlConditionalStringAudio address. Same URL rules as image_url
roleConditionalStringWhat the media is for. Required when type is image_url / video_url / audio_url; unused for type=text

Valid type / role combinations โ€‹

typerolePurpose
textunusedText prompt
image_urlfirst_frameFirst frame
image_urllast_frameLast frame
image_urlreference_imageReference image
video_urlreference_video (fixed)Reference video
audio_urlreference_audio (fixed)Reference audio

role is required for all three media types; omitting it or leaving it empty is rejected.

Reference input conditions โ€‹

ItemCondition
First and last framesAt most 1 each. At least 1 type=text element must be present as well
Ratio of first/last framesDerived from the input image when ratio is omitted; a supported ratio may also be set explicitly
Recommended reference mixUp to 9 images, 3 videos, and 3 audio clips, with no more than 12 items in total
Audio referencesPair audio with at least 1 reference image or reference video; do not submit audio alone
Registered materialsPass asset://<id> as image_url.url; the material must be DONE

Do not apply VOD / MPS specs here

The reference counts above are recommendations for composing Tokenhub requests. Do not treat the FileInfos counts or size limits of VOD / MPS as limits for this API.

Request examples โ€‹

Text-to-video โ€‹

bash
curl -X POST 'https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-videos/tasks' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "wand-vega-video-1.0-pro",
    "content": [
      {"type": "text", "text": "A woman walks down a rain-soaked neon street, cinematic camera move, natural motion."}
    ],
    "resolution": "1080P",
    "ratio": "16:9",
    "duration": 10
  }'

First-and-last-frame generation โ€‹

json
{
  "model": "wand-vega-video-1.0-pro",
  "content": [
    {"type": "text", "text": "Transition naturally from the first frame to the last frame, keeping the lighting continuous."},
    {"type": "image_url", "image_url": {"url": "https://example.com/start.jpg"}, "role": "first_frame"},
    {"type": "image_url", "image_url": {"url": "https://example.com/end.jpg"}, "role": "last_frame"}
  ],
  "resolution": "1080P",
  "duration": 8
}

Reference-based generation (fast variant) โ€‹

json
{
  "model": "wand-vega-video-1.0-pro-fast",
  "content": [
    {"type": "text", "text": "The character from the reference image moves naturally inside the reference video's scene, in rhythm with the music."},
    {"type": "image_url", "image_url": {"url": "https://example.com/character.jpg"}, "role": "reference_image"},
    {"type": "video_url", "video_url": {"url": "https://example.com/scene.mp4"}, "role": "reference_video"},
    {"type": "audio_url", "audio_url": {"url": "https://example.com/bgm.mp3"}, "role": "reference_audio"}
  ],
  "resolution": "720P",
  "ratio": "16:9",
  "duration": -1,
  "options": {"return_last_frame": true}
}

Referencing a registered material โ€‹

Material registration and query are covered in Material management; real-person material additionally requires liveness verification first.

json
{
  "model": "wand-vega-video-1.0-pro",
  "content": [
    {"type": "text", "text": "The subject slowly turns to face the camera as sunlight falls across the frame."},
    {"type": "image_url", "image_url": {"url": "asset://asset-20260914103210-5832"}, "role": "reference_image"}
  ],
  "resolution": "1080P",
  "ratio": "16:9",
  "duration": 5
}

Response parameters โ€‹

A successful submission returns HTTP 201 Created with no error block. Task acceptance does not mean the video is finished.

ParameterTypeDescription
idStringTask ID, used to poll task status
modelStringEcho of the submitted model value
statusStringTask status. Always queued on successful submission
json
{
  "id": "1300000000-video-68b7c1d4a1f2e3d4c5b6a79808192a3b4c5d6e7f8-0000",
  "model": "wand-vega-video-1.0-pro",
  "status": "queued"
}

Querying task results โ€‹

Endpoint: GET https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-videos/tasks/{id}

bash
curl -X GET 'https://tokenhub-intl.tencentcloudmaas.com/v1/wand/vega-videos/tasks/<task_id>' \
  -H 'Authorization: Bearer YOUR_API_KEY'
ParameterTypeDescription
idStringTask ID
modelStringEcho of the submitted model value โ€” not an identifier of the model that actually produced the result
statusStringqueued / running / succeeded / failed
contentObjectGeneration result, returned only when status=succeeded
content.video_urlStringTemporary download address of the finished video, valid for 12 hours
content.last_frame_urlStringLast-frame image address. Returned only when options.return_last_frame=true and the model actually produced a last frame
usageObjectToken usage statistics. Returned only when options.return_vega_usage=true was submitted and status=succeeded
usage.completion_tokensIntegerTokens consumed by output (video generation)
usage.total_tokensIntegerTotal tokens consumed by the task (input + output)
errorObjectFailure details, returned only when status=failed
error.codeStringTask-level error code, e.g. InternalError / TaskTimeout
error.messageStringFailure reason

Handling by status โ€‹

statusMeaningHandling
queuedWaitingKeep querying the same task ID every 3โ€“5 seconds
runningGeneratingKeep querying the same task ID
succeededDoneDownload the result and stop querying
failedFailedRead error to diagnose, then stop querying
json
{
  "id": "1300000000-video-68b7c1d4a1f2e3d4c5b6a79808192a3b4c5d6e7f8-0000",
  "model": "wand-vega-video-1.0-pro",
  "status": "succeeded",
  "content": {
    "video_url": "https://aigc-video.example.com/xxx/result.mp4"
  }
}

Re-host the result before the URL expires. Tasks that time out while waiting return failed with TaskTimeout.

Error codes โ€‹

The error-code vocabulary is shared with the other versions and the material API โ€” see the unified error codes on the overview page.

Built with VitePress ยท Deployed on Tencent EdgeOne Pages