Google's Veo 3.1 prompting guide gives a five-part formula and a separate way to direct sound. This sheet quotes both.
Where Veo 3.1 runs (13 Oct 2026; Gemini API line re-read 3 Nov 2026): in Google Flow, and on Google Cloud’s Gemini Enterprise Agent Platform, where Veo 3.1 and Veo 3.1 Fast are generally available. 1 2 In the Gemini app, video now comes from Gemini Omni, which Google says “will replace Veo in the Gemini app”. 3 On the Gemini API, Google says the three Veo 3.1 preview models (veo-3.1-generate-preview, veo-3.1-fast-generate-preview and veo-3.1-lite-generate-preview) were shut down on 22 Oct 2026, and it names gemini-omni-1.1-flash as the replacement. 45 The prompt advice below is written for Veo 3.1 wherever you use it.
- Formula: [Cinematography] + [Subject] + [Action] + [Context] + [Style & Ambiance].
- Camera first. Google's formula opens with cinematography, not the subject.
- Sound has its own syntax. Dialogue after a colon with no quotation marks (
A woman says: My name is Clara.),SFX:for effects,Ambient noise:for background. - Exclusions: be specific. Google: “describe what you wish to exclude” — its example is “a desolate landscape with no buildings or roads”.
- A
negativePromptfield exists on Gemini Enterprise Agent Platform (formerly Vertex AI), not in the Gemini API.
Every rule on this sheet is quoted from the vendor's own published documentation, linked where it is used.
The formula
Google Cloud, Ultimate prompting guide for Veo 3.1, read at source 16 Sep 2026: “[Cinematography] + [Subject] + [Action] + [Context] + [Style & Ambiance]”
Template:
[shot size and camera move], [subject described specifically], [what they do], [environment and time], [aesthetic, mood, lighting]. [Who] says: [the line, no quotation marks]. SFX: [effect]. Ambient noise: [background].
Worked example:
Medium shot, slow push-in, a ceramicist in her fifties, centring a ball of clay and drawing it upward, north-facing studio at dusk with the wheel between her knees, muted earth palette and window light only, photoreal. SFX: the low hum of a potter's wheel. Ambient noise: quiet room tone, distant traffic.
SFX: the low hum of a potter’s wheel. Ambient noise: quiet room tone, distant traffic.Breakdown — what each part does
Cinematography
Google: “Define the camera work and shot composition.”
Subject
Google: “Identify the main character or focal point.”
Action
Google: “Describe what the subject is doing.”
Context
Google: “Detail the environment and background elements.”
Style & Ambiance
Google: “Specify the overall aesthetic, mood, and lighting.”
Audio
Google: “Veo 3.1 can generate a complete soundtrack based on your text instructions.” SFX: for effects, Ambient noise: for the soundscape.
Dialogue: a colon, no quotation marks. Google’s newest Veo guide, updated 9 Oct 2026, says to “use a colon ( : ) after the speaker’s action to denote speech and avoid using quotation marks”, because quotation marks can make the model write the words into the video as text. Its example: A woman says: My name is Clara. 6
Dated note: Google’s October 2025 prompting guide said the opposite, “Use quotation marks for specific speech”, and its Gemini API page, updated 8 Oct 2026, still says “Dialogue: Use quotes for specific speech.” An earlier version of this sheet followed that advice; it now follows the newer guide, which gives a reason. 7 8
Does it take a negative prompt?
negativePrompt on Gemini Enterprise Agent Platform, formerly Vertex AI. Not in the Gemini API.YESGoogle Cloud, Generate videos from text prompts, read at source 16 Sep 2026: “A string value that describes content that you want to prevent the model from generating.” docs.cloud.google.com. The Gemini API’s Veo reference lists no negative prompt parameter.
In the prompt itself, Google's advice is specificity, not a ban on the word "no". Its example keeps "no": “a desolate landscape with no buildings or roads” instead of “no man-made structures”. Name the exact things that should be absent.
Google Cloud, Ultimate prompting guide for Veo 3.1, read at source 16 Sep 2026: “To refine your output, describe what you wish to exclude.” An earlier version of this sheet told you to write plain nouns only and never "no"; Google's own example uses "no", so that rule was removed.
Open with the camera, label the sound, and name exactly what should be absent.
What not to do
SFX: and Ambient noise:.✕A woman says: My name is Clara. 6✕Free or paid? What the free plan allows, who owns the output, and when you need to pay: free vs paid AI image, video and music tools.
Where this model sits among the others: which models take a negative prompt, and what each vendor publishes as its formula.
What a Veo clip costs is priced by the second, not by the clip.
The checklist
Before you send it:
SFX: and Ambient noise: labelled?□Google Cloud, Ultimate prompting guide for Veo 3.1 · Google Cloud, Generate videos from text prompts All read at source on 16 Sep 2026. The numbered sources below were read on 13 Oct 2026.
Model versions and vendor documentation both move. Re-read the source before relying on a specific number.
Every other sheet in this set: all model cheat sheets.
Related on this site: No Negative Prompt Field? · First Frame and Last Frame in AI Video · Sound in AI Video
- 1Google Cloud, Veo 3.1, Gemini Enterprise Agent Platform documentation, last updated 10 Oct 2026, read at source 13 Oct 2026veo-3.1-generate-001 and veo-3.1-fast-generate-001, “Launch stage: GA”, retirement date “November 17, 2026 or later”; veo-3.1-lite-generate-001, “Launch stage: Preview”. ↑
- 2Google, Learn about Google Flow models & supported features, Google Flow Help, read at source 13 Oct 2026sections “Supported features in Veo 3.1 - Lite”, “Veo 3.1 - Fast”, “Veo 3.1 - Quality” and “Supported features in Gemini Omni Flash 1.1”. ↑
- 3Google, Gemini Omni (Gemini app overview), undated, read at source 13 Oct 2026“Gemini Omni is our latest video editing and generation model that will replace Veo in the Gemini app.” ↑
- 4Google, Gemini API deprecations, last updated 12 Oct 2026, read at source 13 Oct 2026veo-3.1-generate-preview, veo-3.1-fast-generate-preview and veo-3.1-lite-generate-preview, shutdown “October 22, 2026”, recommended replacement “gemini-omni-1.1-flash”; the page notes that its dates “indicate the earliest possible dates on which a model might be retired”. ↑
- 5Google, Video generation in the Gemini API, read at source 3 Nov 2026“Veo 3.1 Preview (veo-3.1-generate-preview, veo-3.1-fast-generate-preview, and veo-3.1-lite-generate-preview) in the Gemini API was shut down on October 22, 2026; migrate to Gemini Omni Flash (gemini-omni-1.1-flash) or the GA models available through the Gemini Enterprise Agent Platform to avoid service interruptions.” ↑
- 6Google Cloud, Best practices for generating videos, Gemini Enterprise Agent Platform documentation (the older Vertex AI address redirects here), last updated 9 Oct 2026, read at source 13 Oct 2026“To prevent the model from rendering text in the video, use a colon ( : ) after the speaker’s action to denote speech and avoid using quotation marks”; its recommended example is “A woman says: My name is Clara.” ↑a“To prevent the model from rendering text in the video, use a colon ( : ) after the speaker’s action to denote speech and avoid using quotation marks”; its recommended example is “A woman says: My name is Clara.” ↑b
- 7Google Cloud Blog, The ultimate prompting guide for Veo 3.1, 16 Oct 2025, read at source 13 Oct 2026“Dialogue: Use quotation marks for specific speech”, with the example A woman says, “We have to leave now.” ↑
- 8Google AI for Developers, Generate videos with Veo 3.1 in Gemini API, last updated 8 Oct 2026, read at source 13 Oct 2026“Dialogue: Use quotes for specific speech.” ↑
Free printable AI cheat sheets, and AI news every week.
Sixteen one-page AI guides and checklists in one free PDF, no sign-up. And if you want to keep up, the newsletter covers what changed in AI each week, in plain English, with the sources.
SUBSCRIBE FREE ↗ GET THE CHEAT SHEET PACK