Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Alibaba Cloud / HappyHorse 1.1

HappyHorse 1.1 Image-to-Video

Animates a single opening image while preserving its aspect ratio, with generated audio.

availableClosed weightsVideo generation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
29 of 36 specification fields collected
Output resolution480p, 720p, 1080pSee specification notes below
Clip duration3–15 secondsSee specification notes below
Input billingSee ratesAlibaba Cloud Model Studio · China (Beijing)
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected
Maximum input tokens
Not collected
Maximum output
Not collected
Output modalities
Video, AudioSource for Output modalities (opens in a new tab)

Audio is synchronized within the generated video, not a standalone text-to-speech or audio-file API.

Reasoning
Not collected
Streaming
Not supportedSource for Streaming (opens in a new tab)

This HTTP API returns a task ID followed by a completed video URL; it does not stream generated frames.

Knowledge cutoff
Not collected
Release date
2026-06-16Source for Release date (opens in a new tab)

International Model Studio availability, distinct from the public product announcement.

Parameters
Not collected
Architecture
Not collected
Languages
Prompts accept any language; no exhaustive dialogue/lip-sync language roster publishedSource for Languages (opens in a new tab)
Output resolutions
480p, 720p, 1080pSource for Output resolutions (opens in a new tab)
Prompt length limit
5,000 non-Chinese characters or 2,500 Chinese charactersSource for Prompt length limit (opens in a new tab)

Excess characters are truncated; these are character limits, not tokens.

Request mode
Asynchronous task creation and pollingSource for Request mode (opens in a new tab)

X-DashScope-Async: enable is required. One output video per task; not a discounted batch or token-streaming API.

Random seed range
0–2,147,483,647Source for Random seed range (opens in a new tab)

Matching seeds improve reproducibility but do not guarantee identical results.

Visible watermark
Optional visible HappyHorse watermark; enabled by defaultSource for Visible watermark (opens in a new tab)
Result and task expiry
Task queries and generated video URLs are valid for 24 hoursSource for Result and task expiry (opens in a new tab)

Download completed files before expiry. This is separate from account, log and policy retention.

Usage metering
Billed generated-video seconds, at the selected output resolutionSource for Usage metering (opens in a new tab)

List prices exclude promotions, subscription credits and other hosting products. Failed requests incur no charge.

Data retention
Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.Source for Data retention (opens in a new tab)
Web product content-use policy
HappyHorse web terms permit model development using non-personal user content; this is separate from Model Studio API policySource for Web product content-use policy (opens in a new tab)

The product terms license user content broadly and reserve model rights. Output ownership does not convey the model weights.

Web product retention
Purpose-based retention with legal exceptions; no fixed general retention periodSource for Web product retention (opens in a new tab)

Applies to the HappyHorse web product, not the region-specific Model Studio endpoints listed here.

Published training data
Text, image, video and audio from public internet, partner, contracted/annotated and synthetic sourcesSource for Published training data (opens in a new tab)

Family-level disclosure says trillions of training tokens; exact corpus sizes and checkpoint mixtures are not published. This is not an inference context limit.

Image input limits
Exactly one first-frame image; 20 MB each; both sides ≥300 px; aspect ratio 1:2.5–2.5:1; JPEG/JPG/PNG/WebPSource for Image input limits (opens in a new tab)
Input media transport
Public HTTP(S) image URL or base64 data URLSource for Input media transport (opens in a new tab)
Aspect ratios
Preserves input image aspect ratio; pixel count approximates the selected resolution tierSource for Aspect ratios (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Alibaba Cloud Model Studio · China (Beijing)

Provider source (opens in a new tab)
API model ID
happyhorse-1.1-i2v
Access
API key, model and endpoint must match the same region. POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll the returned task ID.
Regions
China (Beijing). Deployment/billing scope is not a verified country-only data-residency guarantee.
Published request quota
300 requests/minuteSource for Published request quota (opens in a new tab)

Published request quota, not concurrent generation capacity or guaranteed throughput.

Billing and deployment scope
China (Beijing)Source for Billing and deployment scope (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation480P output video1$0.0618845second of output video
Video generation720P output video1$0.123769second of output video
Video generation1080P output video1$0.165026second of output video

Rate conditions

  1. Original USD list rate for China (Beijing); no currency conversion. Charge = rate × output duration; input is not billed. See the linked billing guide for free-quota eligibility and successful-task billing. Promotional discounts are excluded.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · Singapore

Provider source (opens in a new tab)
API model ID
happyhorse-1.1-i2v
Access
API key, model and endpoint must match the same region. POST https://{WorkspaceId}.ap-southeast-1.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll the returned task ID.
Regions
Singapore / International. Deployment/billing scope is not a verified country-only data-residency guarantee.
Published request quota
300 requests/minuteSource for Published request quota (opens in a new tab)

Published request quota, not concurrent generation capacity or guaranteed throughput.

Billing and deployment scope
InternationalSource for Billing and deployment scope (opens in a new tab)
Included free allowance
10 seconds per model; 90-day eligibility windowSource for Included free allowance (opens in a new tab)

Window begins at the latest of Model Studio activation, model release or access approval. Other listed regions have no free quota; this is not a permanent free tier.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation480P output video1$0.07second of output video
Video generation720P output video1$0.14second of output video
Video generation1080P output video1$0.18second of output video

Rate conditions

  1. Original USD list rate for Singapore / International; no currency conversion. Charge = rate × output duration; input is not billed. See the linked billing guide for free-quota eligibility and successful-task billing. Promotional discounts are excluded.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · Germany (Frankfurt)

Provider source (opens in a new tab)
API model ID
happyhorse-1.1-i2v
Access
API key, model and endpoint must match the same region. POST https://{WorkspaceId}.eu-central-1.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll the returned task ID.
Regions
Germany (Frankfurt) / Global. Deployment/billing scope is not a verified country-only data-residency guarantee.
Published request quota
300 requests/minuteSource for Published request quota (opens in a new tab)

Published request quota, not concurrent generation capacity or guaranteed throughput.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation480P output video1$0.0618845second of output video
Video generation720P output video1$0.123769second of output video
Video generation1080P output video1$0.165026second of output video

Rate conditions

  1. Original USD list rate for Germany (Frankfurt) / Global; no currency conversion. Charge = rate × output duration; input is not billed. See the linked billing guide for free-quota eligibility and successful-task billing. Promotional discounts are excluded.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · US (Virginia)

Provider source (opens in a new tab)
API model ID
happyhorse-1.1-i2v
Access
API key, model and endpoint must match the same region. POST https://{WorkspaceId}.us-east-1.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll the returned task ID.
Regions
US (Virginia) / Global. Deployment/billing scope is not a verified country-only data-residency guarantee.
Published request quota
300 requests/minuteSource for Published request quota (opens in a new tab)

Published request quota, not concurrent generation capacity or guaranteed throughput.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation480P output video1$0.0618845second of output video
Video generation720P output video1$0.123769second of output video
Video generation1080P output video1$0.165026second of output video

Rate conditions

  1. Original USD list rate for US (Virginia) / Global; no currency conversion. Charge = rate × output duration; input is not billed. See the linked billing guide for free-quota eligibility and successful-task billing. Promotional discounts are excluded.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.

Read the service policy (opens in a new tab)

Alibaba Cloud Model Studio · Japan (Tokyo)

Provider source (opens in a new tab)
API model ID
happyhorse-1.1-i2v
Access
API key, model and endpoint must match the same region. POST https://{WorkspaceId}.ap-northeast-1.maas.aliyuncs.com/api/v1/services/aigc/video-generation/video-synthesis; poll the returned task ID.
Regions
Japan (Tokyo) / Global. Deployment/billing scope is not a verified country-only data-residency guarantee.
Published request quota
300 requests/minuteSource for Published request quota (opens in a new tab)

Published request quota, not concurrent generation capacity or guaranteed throughput.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Video generation480P output video1$0.0618845second of output video
Video generation720P output video1$0.123769second of output video
Video generation1080P output video1$0.165026second of output video

Rate conditions

  1. Original USD list rate for Japan (Tokyo) / Global; no currency conversion. Charge = rate × output duration; input is not billed. See the linked billing guide for free-quota eligibility and successful-task billing. Promotional discounts are excluded.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard inference retention is up to 30 days, with legal/security exceptions; stateful files persist until deleted. HappyHorse is absent from the ZDR-supported model list, and asynchronous task paths are explicitly excluded. The training-data disclosure requires explicit consent to use customer business data for model improvement. Expiring video URLs are not a blanket zero-retention promise.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Alibaba Cloud Model Studio · China (Beijing)

happyhorse-1.1-i2v
Not supported

The model card explicitly marks context caching unsupported. Temporary output storage is not discounted prompt caching.

Support evidence (opens in a new tab)Caching reviewed 14 Oct 2026

Alibaba Cloud Model Studio · Singapore

happyhorse-1.1-i2v
Not supported

The model card explicitly marks context caching unsupported. Temporary output storage is not discounted prompt caching.

Support evidence (opens in a new tab)Caching reviewed 14 Oct 2026

Alibaba Cloud Model Studio · Germany (Frankfurt)

happyhorse-1.1-i2v
Not supported

The model card explicitly marks context caching unsupported. Temporary output storage is not discounted prompt caching.

Support evidence (opens in a new tab)Caching reviewed 14 Oct 2026

Alibaba Cloud Model Studio · US (Virginia)

happyhorse-1.1-i2v
Not supported

The model card explicitly marks context caching unsupported. Temporary output storage is not discounted prompt caching.

Support evidence (opens in a new tab)Caching reviewed 14 Oct 2026

Alibaba Cloud Model Studio · Japan (Tokyo)

happyhorse-1.1-i2v
Not supported

The model card explicitly marks context caching unsupported. Temporary output storage is not discounted prompt caching.

Support evidence (opens in a new tab)Caching reviewed 14 Oct 2026

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

HappyHorse’s official terms reserve rights in its models, grant only limited service-use rights and prohibit extracting or reverse-engineering the underlying parameters and weights. Alibaba identifies HappyHorse as its own-developed model, distributed through the official product and Model Studio API; no separate public-weight license or release is documented in the reviewed official sources. This classification does not extend to the separate Qwen or Wan families, and API wrappers are not model-weight releases.

Classification evidence (opens in a new tab)

Lifecycle dates

International Model Studio release16 Jun 2026Source (opens in a new tab)
Provider republishes the 1.1 upgrade announcement23 Jun 2026Source (opens in a new tab)

Sources reviewed