Googles nyeste videomodell, Veo 3, tilbyr forbedret kvalitet i forhold til Veo 2. Med den kan du nå legge til dialog mellom karakterer, lydeffekter og bakgrunnsstøy for en rikere seeropplevelse.
curl-X POST 'https://pollo.ai/api/platform/v1/generation/google/veo-3/video'\-H'Content-Type: application/json'\-H'x-api-key: YOUR_API_KEY'\-d'{ "input": { "prompt": "A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field", "negativePrompt": "blurry, low quality, distorted, watermark, text", "aspectRatio": "16:9", "duration": 4, "resolution": "720p", "generateAudio": true }}'
Du kan eventuelt inkludere webhookUrl for å motta et tilbakekall når oppgaven lykkes eller mislykkes.
Valgfritt forespørselsfelt
{"input":{"prompt":"A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field","negativePrompt":"blurry, low quality, distorted, watermark, text","aspectRatio":"16:9","duration":4,"resolution":"720p","generateAudio":true},"webhookUrl":"https://example.com/webhooks/pollo"}
Skjema
Inndatafelt for det valgte endepunktet og modusen.
Felt
Type
Obligatorisk
Beskrivelse
prompt
string
Ja
Text prompt describing the content to generate.
negativePrompt
string
Ingen
Text describing elements to avoid in the generated output.
aspectRatio
string
Ingen
Output aspect ratio, as width:height (e.g. 16:9).
duration
number
Ingen
The duration of the generated video, in seconds.
resolution
string
Ingen
Output resolution of the generated media (e.g. 720p, 1080p, 2K, 4K). Values are case-insensitive.
seed
integer
Ingen
Random seed for reproducible generation.
generateAudio
boolean
Ingen
Generate natural-sounding, fitting audio for the output video.
Audio is part of the generation plan in Veo 3. The model can create short video clips where speech, sound effects, ambient noise, motion, and camera direction are described together.
With Veo 3 on Pollo API, developers can add text-to-video and image-to-video workflows to products that need generated clips with both visual and audio output.
Key Features of Veo 3 API
Native Audio Output
Veo 3 can produce generated audio with the video, including speech, scene ambience, and sound effects tied to visible actions in the clip.
Short Dialogue Handling
The model is useful for brief spoken moments where the prompt defines the speaker, line, gesture, and surrounding reaction in a compact scene.
Motion and Physical Detail
Veo 3 handles object movement, lighting changes, camera motion, and environmental effects in ways that help short scenes feel internally consistent.
Text and Image Guidance
Prompts can start from text alone or use an image as visual context for the subject, opening frame, product, object, or environment.
Sound-Aware Scene Design
Scene instructions can include what should be heard, such as footsteps, tools, alarms, room tone, water, machinery, or a short spoken line.
Compact Clip Structure
Veo 3 is easier to control when the scene has one main action, one clear camera idea, and an audible or visible ending beat.
Use Cases of Veo 3 API
Dialogue Clip Tools: Generate short scenes with a defined speaker, spoken line, matching action, and surrounding ambience.
Training Scenarios: Convert workplace, safety, or onboarding scripts into short clips with visible actions and relevant audio cues.
Educational Demonstrations: Create lesson clips where narration, object movement, and environmental sound support a single teaching point.
Game and Story Moments: Produce brief cutscenes, NPC lines, or branching story assets that need voice, motion, and setting detail together.
Ad Concept Drafts: Test short commercial scenes before production, especially when sound, action, and timing affect the idea.
Creator App Features: Let users create short video posts with generated sound instead of sending silent clips to a separate audio workflow.
Image-to-Video Animation: Animate uploaded visuals by defining the camera move, subject action, environment, and sound treatment.
How to Use Veo 3 API
Get an API Key: Create a Pollo API account and generate your API key from the developer dashboard.
Choose the Model: Call the Veo 3 route at /generation/google/veo3.
Add Your Inputs: Send a prompt for text-to-video or an image URL for image-to-video; optionally set negativePrompt, length, aspectRatio, resolution, seed, or generateAudio.
Generate and Retrieve: Submit the task, save the returned task information, and retrieve the video after generation completes.
Prompting Best Practices for Veo 3 API
Write prompts as short scene specifications. Include the subject, setting, camera position, main action, audio requirement, spoken line if needed, and final moment.
A Simple Prompt Formula
Subject + setting + camera position + main action + audio cue or dialogue + visual detail + ending beat
What You Should Notice
Audio Scope: State whether the clip should include speech, ambience, sound effects, silence, or a specific combination.
Dialogue Length: Keep spoken lines short, especially when the scene also needs visible action or a reaction.
Action-Sound Link: Describe actions that naturally create sound, such as a door opening, a tool turning, water pouring, or a timer beeping.
Camera Simplicity: Use one clear camera idea instead of packing several shot changes into a short output.
Image Purpose: If an image is provided, explain whether it defines the subject, environment, first frame, object shape, or style reference.
End State: Define what should be true at the end of the clip, such as the object fixed, the door open, or the speaker paused.
Example Prompts
Transit Platform Announcement
"Create an 8-second video on an early morning train platform. A station worker checks a digital board, points passengers toward another track, and says, 'Platform three is now boarding.' Include PA reverb, rolling luggage, cool light, and a slow side-tracking camera."
Workshop Repair Clip
"Use the uploaded photo of a broken desk lamp as the starting image. Show hands replacing the switch, the lamp flickering once, then glowing steadily. Include small screwdriver sounds, quiet workbench ambience, warm shadows, and a close final push-in."
Classroom Chemistry Demo
"Create a short classroom demonstration. A teacher adds blue solution to a clear beaker, students lean in, and the liquid turns green. Add the line 'Watch the reaction start,' glass clinks, room tone, and a steady overhead-to-close camera move."
Emergency Drill Scenario
"Generate a workplace safety drill clip. A hallway light flashes, an employee closes a cabinet, follows floor arrows, and says, 'Exit route is clear.' Include a distant alarm, footsteps, fluorescent lighting, and a final shot of the open stairwell door."
Cooking Timer Moment
"Create an 8-second kitchen clip. A cook opens the oven, steam escapes, a timer beeps twice, and they say, 'Perfect, let it rest.' Use close handheld framing, warm counter light, oven sounds, and a final pause on the tray."
Veo 3 vs Veo 3 Fast vs Veo 3.1
Capability
Veo 3
Veo 3 Fast
Veo 3.1
Main Use
Standard Veo 3 generation with video and native audio
Speed-oriented iteration for drafts and prompt testing
More recent Veo workflow with expanded visual control
Input Fit
Text prompts and single-image guidance
Text and image prompts for quick variants
Text, images, references, and controlled frame workflows
Audio Workflow
Generated speech, ambience, and sound effects in short clips
Useful for testing audio-video prompt ideas quickly
Useful when audio needs to follow more controlled visual inputs
Scene Control
Concise scenes with one main action and a defined sound cue
Multiple prompt directions and variant exploration
Reference-led or frame-guided scene planning
Developer Fit
Apps that need generated video with audio in the same workflow
Preview tools, batch ideation, and short iteration loops
Workflows that need tighter visual continuity or reference handling
Recommended For
Dialogue clips, explainers, training scenes, and image-to-video outputs
Drafting, testing, and prompt comparison
Campaign, reference, or storyboard assets with stricter visual control
Why Choose Veo 3 API?
Veo 3 is a practical choice when the output should include generated sound, not only moving images. It suits short scenes with speech, ambience, or action-based audio.
Through Pollo API, teams can access Veo 3 alongside 300+ image and video models, with one API surface, clean documentation, lower costs, scalable infrastructure, and human support.
Use Veo 3 on Pollo API when your product needs audio-aware video generation for dialogue clips, training scenes, education content, ad drafts, or image-to-video tools.
Veo 3 API FAQs
What is Veo 3?
Veo 3 is Google's video generation model for creating short videos with native audio, including speech, sound effects, ambience, realistic motion, and text or image guidance.
Does Veo 3 API support text-to-video and image-to-video?
Yes. Veo 3 API on Pollo API supports prompt-based generation and image-guided generation, so developers can create clips from written scenes or uploaded visuals.
Can Veo 3 generate audio with the video?
Yes. Veo 3 can generate audio with the clip, including short speech, background ambience, and sound effects that relate to visible scene actions.
Is Veo 3 API useful for dialogue scenes?
Yes, especially for brief dialogue. The prompt should define who speaks, the line, the visible action, and the surrounding sound environment.
How is Veo 3 different from Veo 3 Fast?
Veo 3 is better suited to standard generation where audio timing and visual detail matter. Veo 3 Fast is more useful for quick variants and prompt testing.
Why run Veo 3 through Pollo API?
Pollo API lets developers use Veo 3 alongside 300+ other image and video models, making comparison, workflow design, and integration management simpler.
