Grok Video 1.5
Grok Video 1.5 is a video model family by xAI. Use Grok for text, first-frame and image-reference video, or for editing an existing clip. The model author also documents other video operations; only the modes in Vini’s generated table are connected here.
Resolution, duration and sound
Section titled “Resolution, duration and sound”Text and first-frame generation support 480p, 720p and 1080p, with a length of 1–15 seconds. Reference generation supports 480p or 720p, also for 1–15 seconds. Editing supports 480p or 720p and follows the source within its separate 1–8 second input limit.
New nodes use 480p with native sound. The connected API always generates sound and offers no sound-off control. An explicitly saved resolution remains selected. Available resolutions depend on the operation, so 1080p is not offered for references or editing.
Settings were checked against the Grok 1.5 generation API, reference API and edit API on 20 October 2026. This is a settings check, not a new output-quality test.
| Setting | Value |
|---|---|
| Model ID | grok15 |
| Name | Grok Video 1.5 |
| Availability | available |
| Default for new nodes | no |
| Earlier IDs resolving to this record | grok_v1_edit |
Common declared limits
Section titled “Common declared limits”| Setting | Value |
|---|---|
| Capabilities | text to video, image to video, dialogue |
| Maximum prompt length | 4096 |
| Duration | 1–15 seconds; fallback 6 seconds |
| Audio | model-generated audio; fallback model-generated audio |
| Aspect ratio | 9:16, 16:9, 4:3, 3:4, 3:2, 2:3, 1:1; fallback 9:16 |
| Declared language support | prompt: English; speech: English, Russian; tier: platinum |
| Controls | resolution: 480p, 720p, 1080p; fallback 480p |
| Element input form | sheet |
| Content constraints | people: restriction level: ok |
| Negative prompt strategy | positive |
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 |
|---|---|
| Controls | resolution: 480p, 720p; fallback 480p |
| References | maximum: 7 |
Mode: edit_video
Section titled “Mode: edit_video”| Setting | Value |
|---|---|
| Prompt required | yes |
| Duration | same as source video; fallback same as source video |
| Aspect ratio | not declared |
| Combined reference limits | images: 0; videos: 1; audio inputs: 0; combined maximum: 1; audio without visual input: no; video reference seconds: minimum: 1; maximum: 8 |
| Controls | resolution: 480p, 720p; fallback 480p |
| Negative prompt strategy | instruction in prompt text |
| Source video | required |
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”- Write a compact scene description with one clear action and a separately named camera movement.
- Identify the reference subject by visible features instead of relying on a name alone.
- Write exact dialogue only for the person who should speak. For narration, use a separate voice track.
The built-in Grok Video 1.5 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 Grok Video 1.5 and a first frame showing a ceramic mug on a workbench. Ask Prompt for the model: “A hand slowly turns the mug to reveal the painted bird. The camera remains beside the bench. Quiet room ambience.” Review the intended action before generating.
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”Image-reference identity evidence does not establish accurate video-style transfer. A short Russian-speaking trial produced recognizable words, but its actor gestures were imperfect. Avoid describing that single result as general multilingual or acting reliability.
Evidence and limits
Section titled “Evidence and limits”The older reference-mode exam used individual frames. The current sheet input has an outstanding replacement exam; that older success is historical.
The 28 September 2026 reference test retained the fictional character’s clothing and distinctive features. On 11 October, one authoring-adapter trial produced Russian speech and kept the narrator out of the character dialogue. There was no matched writer-without-adapter control, and separate narration assembly was not tested.
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.