VEO 3.1 Extend Video is Google's video extend & retake model on Versely. This page is its structured prompting reference: the 7 parameters its schema actually exposes, the extend-retake technique that applies to it, copy-ready templates.
Everything here is grounded in the same sources Versely's agent reads — the model's input schema, the veo family rule its prompt enhancer applies. Where a line is general craft advice rather than a documented fact about VEO 3.1 Extend Video, the page says so.
What VEO 3.1 Extend Video wants
The exact input surface, from the same schema the Versely agent fetches with get_model_input_schema before every generation.
| Parameter | What it does | Values |
|---|---|---|
video_urlreq | Source VEO 3.1 video to extend | string |
promptreq | Continuation prompt (max 20000 chars) | string (max 20000) |
duration | CONST "7s" — the OpenAPI schema shows a free string, but the live endpoint validates Literal['7s'] (422 "Input should be '7s'" reproduced 2026-07-31). Each extend adds a fixed 7s; "up to 30s" refers to total output length. | 7sdefault: 7s |
resolution | Default 720p; input video must be 720p or 1080p | stringdefault: 720p |
aspect_ratio | Aspect ratio | auto · 16:9 · 9:16default: auto |
negative_prompt | Exclusions | string |
generate_audio | Toggle audio | booleandefault: true |
- — duration is Literal['7s'] at runtime despite the free-string OpenAPI schema (422 reproduced 2026-07-31) — each extend adds a fixed 7s, chain calls to go longer (output caps at 30s total); input video must be 720p/1080p in 16:9/9:16; SynthID watermark cannot be disabled
The rule Versely's enhancer applies
Versely's prompt enhancer carries a per-family rule for veo models, applied automatically whenever it rewrites a prompt for VEO 3.1 Extend Video. Verbatim:
“VEO video models respond to cinematic descriptions. Include camera angles, movement types (dolly, crane, steadicam), lighting setup, and temporal flow (slow motion, time-lapse). Be specific about scene transitions.”
Technique that applies here
Extend/retake: continuing or re-rolling an existing generated clip
- This family covers two different operations, and knowing which one you're on changes what the prompt should say: continuing past the end of a clip (Flux 3 Extend Video, Grok Imagine Extend, VEO 3.1 Extend Video — all take a 'what happens next' prompt) versus redoing a segment inside an existing clip (LTX 2.3 Retake Video — takes an edit instruction, per the schema's own example, 'Remove the man').
- VEO 3.1 Extend Video's duration is locked to exactly '7s' per call — the schema note explains the OpenAPI spec shows a free string but the live endpoint only validates the literal '7s' (a 422 error otherwise, reproduced 2026-07-31); 'up to 30s' describes the total you reach by chaining multiple extend calls, not a single request. Write each extend prompt for one 7-second beat, then issue a fresh extend call — with a fresh prompt — for the next one.
- Flux 3 Extend Video doesn't have a resolved parameter schema in our system yet — only duration (5s–20s in 1s steps), resolution (720p/1080p), and aspect_ratio are confirmed from the catalog data, alongside a video-input field. Until a verified schema exists, don't assume it shares Grok's taskId-only restriction or VEO's fixed-7s-per-call behavior; lean on the product's flux family guidance instead — MODEL_TIPS says flux models 'excel with natural language descriptions... specific about composition, color palette, and mood... include camera details like lens type, aperture, and film stock' — which the prompt enhancer applies to any model name containing 'flux.'
- VEO 3.1 Extend Video also gets the product's veo family MODEL_TIPS applied by the enhancer: 'respond to cinematic descriptions. Include camera angles, movement types (dolly, crane, steadicam), lighting setup, and temporal flow... be specific about scene transitions' — useful detail for describing the next beat, since the continuation prompt field itself allows up to 20000 characters.
Copy-ready templates
Replace the bracketed slots; each template says when it's the right shape.
[WHAT HAPPENS NEXT], camera [dolly in / crane up / steadicam follow / static hold], [LIGHTING CONTINUITY NOTE matching the source clip]
Use when: Extending a VEO 3.1 clip one 7-second beat at a time (duration is locked to '7s' per call).
Exclude: [WHAT SHOULD NOT CARRY INTO THE NEW FOOTAGE — e.g. camera shake, text overlays, extra people]
Use when: Using VEO 3.1 Extend Video's optional negative_prompt field to keep an unwanted element from continuing into the extension.
How the Versely agent does this automatically
You can use this page by hand, or let the agent apply the same knowledge. Four real mechanisms — no more, no less:
get_model_input_schema— before generating, the agent looks up VEO 3.1 Extend Video's exact input fields, required fields, allowed values, defaults, and min/max bounds. The parameter table above is that same surface.- The prompt enhancer's family rules — 12 per-family rewrite rules, including the veo rule quoted on this page, shape how a rough prompt gets rewritten.
- The per-provider speech guide — for TTS scripts, the agent follows a provider-specific tag scheme — not relevant to this model, but it's why voiceover scripts come out marked up correctly.
expand_movie_scene— in movie flows, brief scene ideas are rewritten into detailed cinematic descriptions before generation.
Mistakes that waste generations
- Feeding Grok Imagine Extend an uploaded or external video URL — the schema only accepts a prior Grok Imagine task_id, and external URLs are rejected per the docs note.
- Requesting a 10s or 15s VEO 3.1 extend in one call — the live endpoint only validates the literal '7s'; asking for anything else returns a 422 even though the field looks like a free string in the OpenAPI spec.
- Writing an LTX Retake prompt like a continuation ('then she walks off screen') — the field is an edit instruction over an existing segment defined by start_time and duration, not a description of new footage after the clip ends.
The long-form guide
This page is the structured reference. For the essay treatment — worked examples, failure modes, and narrative — read VEO 3.1 Prompting: The Complete Guide.
Frequently asked questions
How long can a VEO 3.1 Extend Video prompt be?+
VEO 3.1 Extend Video's schema caps the prompt at 20,000 characters. Spend them on the subject, motion, and the register its family rewards rather than restating parameters the schema already controls.
Does VEO 3.1 Extend Video support negative prompts?+
Yes — the schema exposes negative_prompt. Put exclusions there instead of writing "no text, no watermark" into the main prompt.
Which aspect ratios does VEO 3.1 Extend Video support?+
The aspect_ratio parameter is an enum: auto, 16:9, 9:16. Set the parameter — describing the frame shape in prose does nothing on its own.
How long can a VEO 3.1 Extend Video generation be?+
Duration is a hard enum: 7s. Write one continuous beat sized to the window you pick, not a script the model will compress.
How does the Versely agent know VEO 3.1 Extend Video's parameters?+
Before generating, the agent calls its get_model_input_schema tool, which looks up the exact input fields, required fields, allowed values, defaults, and min/max bounds for the model; separately, the prompt enhancer applies the veo family rule quoted on this page to the prompt text itself. Nothing on this page is guessed — it is the same schema surface those tools read.
Related prompting guides
Generate with VEO 3.1 Extend Video
VEO 3.1 Extend Video is live in Versely — paste a template above, or just describe what you want and let the agent map it onto the schema for you.