HappyHorse 1.1
HappyHorse 1.1 is a video model family by Alibaba. HappyHorse creates video with text, a first frame or image references, and can edit an existing clip in Vini. Its reference input is useful for naming distinct subjects; the number of images is not a measured number of reliably animated characters.
Resolution and sound
Section titled “Resolution and sound”Choose 720p or 1080p for generation or editing. New nodes use 720p, the lower-cost choice, with sound enabled. Saved explicit settings remain selected.
Generation accepts 3–15 seconds and includes native sound. The connected generation API has no sound-off switch. Editing follows the source length and preserves its sound; it does not expose a sound-off or guaranteed new-sound option.
Settings were checked against the HappyHorse 1.1 API and editing API on 20 October 2026. This establishes the controls available through this connection, not a new acting or speech result.
| Setting | Value |
|---|---|
| Model ID | happyhorse11 |
| Name | HappyHorse 1.1 |
| Availability | available |
| Default for new nodes | no |
| Earlier IDs resolving to this record | happyhorse_edit |
Common declared limits
Section titled “Common declared limits”| Setting | Value |
|---|---|
| Capabilities | text to video, image to video, dialogue |
| Maximum prompt length | 2500 |
| Duration | 3–15 seconds; fallback 5 seconds |
| Audio | model-generated audio; fallback model-generated audio |
| Aspect ratio | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, 9:21, 5:4, 4:5; fallback 9:16 |
| Declared language support | prompt: English; speech: English; tier: gold |
| Controls | resolution: 720p, 1080p; fallback 720p |
| Element input form | sheet |
| Content constraints | people: restriction level: unknown |
| Negative prompt strategy | instruction in prompt text |
Mode: from_first_frame
Section titled “Mode: from_first_frame”| Setting | Value |
|---|---|
| Aspect ratio | not declared |
| First frame | required |
Mode: text_to_video
Section titled “Mode: text_to_video”Uses the common declared limits above.
Mode: from_references
Section titled “Mode: from_references”| Setting | Value |
|---|---|
| References | maximum: 9 |
Mode: edit_video
Section titled “Mode: edit_video”| Setting | Value |
|---|---|
| Prompt required | yes |
| Duration | same as source video; fallback same as source video |
| Audio | origin; fallback origin |
| Aspect ratio | same as source video |
| Combined reference limits | images: 5; videos: 1; audio inputs: 0; combined maximum: 6; audio without visual input: no; video reference seconds: minimum: 3; maximum: 15 |
| Source video | required |
| References | maximum: 5 |
Unavailable modes: avatar_audio, motion_transfer, recast.
Limits apply together. initial is the studio choice for a new node; default is the fallback for an existing node without that value. Declared language or input support does not guarantee a successful result.
Prompting playbook
Section titled “Prompting playbook”- Describe each subject with a visible identity cue and give each person a separate action.
- For editing, state the replacement and the source features to preserve.
- Use a brief spoken line for a first speech test. Listen to pronunciation rather than judging only the transcript.
The built-in HappyHorse 1.1 authoring skill adapts a scene request for this family through Prompt for the model. It prepares instructions; it does not render video or certify the result.
Studio example
Section titled “Studio example”Choose HappyHorse 1.1 in reference mode with two fictional character Elements. Ask Prompt for the model: “The person with the striped scarf gives the blue notebook to the person wearing round glasses. A single medium shot.” Inspect which person moves each object.
A run may consume credits. Review the prompt and inputs, then inspect the finished clip for action, identity, framing and sound. For automation, use the supported MCP workflow.
Known failure modes
Section titled “Known failure modes”A test can retain appearance while missing natural acting. The 11 October Russian speech trial mispronounced a word despite a mostly plausible transcript. An edit passing with clean frames does not establish the same result with a newly changed sheet input.
Evidence and limits
Section titled “Evidence and limits”The older reference and editing exams used individual frames. Both operations now use sheets, and their replacement exams are outstanding; those successes are historical.
On 28 September 2026, the 1.1 reference case retained the fictional character, and the replacement edit retained the street and source movement. The edit request did not expose a separate version identifier, so it is evidence for that dated edit connection, not a version-wide claim. On 11 October, one adapter-authored clip produced Russian speech with a pronunciation problem. Those observations replace older “never tested” notes, but do not provide a broad speech benchmark.
The studio registry establishes available inputs and modes. A recorded render establishes only its observed outcome. The applicability guide explains the distinction. See also character identity, static camera and video style samples.
See the model–technique applicability matrix for connected operations, dated outcomes and configurations that remain untested.
Reviewed 14 October 2026 against the current studio contract and dated studio research. Official model-author documentation checked on that date; author capabilities may exceed Vini’s connected controls. Recheck this playbook when the model version, quality or mode changes.