Skip to content

Hailuo โ€‹

Updated

Endpoint: POST https://mps.intl.tencentcloudapi.comAction: CreateAigcVideoTask

Hailuo is a video generation model that supports text-to-video and first-frame / first-and-last-frame video generation. H3 supports multimodal reference generation that mixes images, video, and audio (up to 9 images / 3 videos / 3 audio clips, 12 in total); the 02 and 2.3 series are simple generation engines with fixed durations.

Basic Information โ€‹

ItemValue
ModelNameHailuo
ModelVersionH3 / 2.3 / 2.3-fast / 02
Guardrail release supportOnly H3 supported

Supported Specifications by Version โ€‹

VersionResolutionAspect RatioLength
H3Default output: 768P
Upscale output: 1080P / 2K / 4K
21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:164โ€“15 seconds
H3-MaxDefault output: 480P / 768P
Upscale output: 1080P / 2K / 4K
16:9 / 9:16 / 4:3 / 3:4 / 1:1 / adaptive5โ€“15 seconds
02 / 2.3Refer to the output options for each versionโ€”6 seconds
10 seconds

Input Conditions โ€‹

VersionCondition
H3Input: reference image 9 images / video 3 / audio 3, total 12; separate from first/last frame mode
H3-MaxInput: text or first/last frame
02 / 2.3Input: H3 multimodal limits do not apply

The 02 and 2.3 series are simple generation engines with a fixed length, while H3 is a multimodal engine that references image, video, and audio together.

Request Parameters โ€‹

ParameterRequiredTypeDescription
ModelNameRequiredStringFixed value Hailuo
ModelVersionOptionalStringH3 / 2.3 / 2.3-fast / 02
PromptRequiredStringGeneration prompt. See the version-specific limits below
NegativePromptOptionalStringElements to exclude from the result. Applied on supported models
EnhancePromptOptionalBooleanAutomatic prompt correction. Default value false
ImageUrlOptionalStringFirst frame image URL. Must be externally accessible and 10MB or less
LastImageUrlOptionalStringLast frame image URL. Used together with ImageUrl
ImageInfos.NOptionalArrayReference image array. Category / Url / ReferenceType, etc.
DurationOptionalIntegerGeneration length (seconds). Only values allowed by each model are applied
ExtraParametersOptionalObjectResolution / AspectRatio / EnableAudio
AdditionalParametersOptionalStringJSON string containing model-specific options. For the seed passing format, refer to Additional Configuration
ExtraParameters.
EnablelnputComplianceCheck
OptionalBooleanIf false is explicitly passed, input moderation is released. Contact your sales representative before use
ExtraParameters.
EnableOutputComplianceCheck
OptionalBooleanIf false is explicitly passed, output moderation is released. Contact your sales representative before use

Request Example โ€‹

text-to-video (H3) โ€‹

H3 reflects the camera movement described in the prompt. The default output resolution is 2560x1440.

json
{
  "ModelName": "Hailuo",
  "ModelVersion": "H3",
  "Prompt": "a girl running toward the camera on a playground holding a kite, orbit camera, cinematic quality",
  "Operator": "wiki-example",
  "StoreCosParam": {
    "CosBucketName": "<bucket-appid>",
    "CosBucketRegion": "<bucket-region>",
    "CosBucketPath": "/aigc/"
  },
  "EnhancePrompt": false
}

H3-Max / First/Last Frame โ€‹

json
{
  "ModelName": "Hailuo",
  "ModelVersion": "H3-Max",
  "Prompt": "The camera moves from the first frame toward the final composition.",
  "NegativePrompt": "blur, distortion",
  "EnhancePrompt": false,
  "ImageUrl": "https://<host>/first.jpg",
  "LastImageUrl": "https://<host>/last.jpg",
  "Duration": 5,
  "ExtraParameters": {
    "Resolution": "768P",
    "AspectRatio": "16:9"
  },
  "StoreCosParam": {
    "CosBucketName": "<bucket-appid>",
    "CosBucketRegion": "<bucket-region>",
    "CosBucketPath": "/aigc/"
  },
  "Operator": "wiki-example"
}

Response Example โ€‹

json
{
  "Response": {
    "Status": "DONE",
    "Message": "ok",
    "Resolution": "1920x1080",
    "VideoUrls": [
      "https://aigc-output-video-<id>.cos.<region>.myqcloud.com/<file>.mp4?q-sign-algorithm=sha1&..."
    ],
    "RequestId": "<request-id>"
  }
}

Special Configuration โ€‹

Guardrail Release โ€‹

Hailuo supports guardrail release only in ModelVersion=H3. Do not apply the same setting to other versions.

To use the guardrail release parameters, you must first contact your sales representative to confirm usage permission. Guardrail release is applied only when the parameter is explicitly passed, even on supported engines. If the parameter is omitted, the guardrail is not released.

ParameterRelease ValueScope
ExtraParameters.EnablelnputComplianceCheckfalseInput moderation release
ExtraParameters.EnableOutputComplianceCheckfalseOutput moderation release

Merge the following settings into your existing generation request. To release both input and output moderation, pass both fields together.

json
{
  "ExtraParameters": {
    "EnablelnputComplianceCheck": false,
    "EnableOutputComplianceCheck": false
  }
}

Supported Capabilities by Version โ€‹

VersionGeneration ModeLengthRemarks
H3text / image / multimodal reference4~15 secondsReferences image 9, video 3, audio 3. Native voice generation
2.3text / image6 seconds, 10 seconds1080P is 6 seconds only
2.3-fastimage only6 seconds, 10 secondsFirst frame required
02text / image, first + last frame6 seconds, 10 seconds16:9 only

H3 Input Constraints โ€‹

Image input (first frame, last frame, reference image)

ItemConstraint
FormatJPG, JPEG, PNG, WEBP, HEIC, HEIF
File size30MB or less
Width and height range256~5760 px
Aspect ratio0.4~2.5
CountFirst frame 1 image, last frame 1 image, reference images 9 images or fewer

Video input (reference generation)

ItemConstraint
ContainerMP4, MOV
CodecVideo H.264, H.265 / Audio AAC, MP3
File size50MB or less
Count3 or fewer
Segment length2~15 seconds, total 15 seconds or less
Width and height range256~5760 px
Aspect ratio0.4~2.5
Frame rate23.976~60

Audio input (reference generation)

ItemConstraint
FormatWAV, MP3
File size15MB or less
Count3 or fewer
Segment length2~15 seconds, total 15 seconds or less

image-to-video (First / Last Frame) โ€‹

Enter ImageUrl and LastImageUrl, 1 image each. First/last frame (i2va) and multimodal reference (r2va) are selected as separate generation modes.

json
{
  "ModelName": "Hailuo",
  "ModelVersion": "H3",
  "Prompt": "camera slowly pushes in",
  "ImageUrl": "https://<cdn>/first.jpg",
  "LastImageUrl": "https://<cdn>/last.jpg",
  "Duration": 6,
  "ExtraParameters": {
    "Resolution": "1080P",
    "AspectRatio": "9:16"
  },
  "Operator": "admin",
  "StoreCosParam": {
    "CosBucketName": "<bucket-appid>",
    "CosBucketRegion": "<bucket-region>",
    "CosBucketPath": "/aigc/"
  },
  "EnhancePrompt": false
}

Additional Options โ€‹

i2va and r2va are mutually exclusive

Select the r2va mode for reference video / audio generation, and the i2va mode for first/last frame generation.

Built with VitePress ยท Deployed on Tencent EdgeOne Pages