curl-X POST 'https://pollo.ai/api/platform/v1/generation/vidu/viduq4/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", "image": "https://example.com/first-frame.jpg", "duration": 5, "resolution": "540p", "generateAudio": true }}'
Optionally include webhookUrl to receive a callback when the task succeeds or fails.
Optional request field
{"input":{"prompt":"A cinematic shot of a golden retriever running through a field of sunflowers at sunset, warm rim light, shallow depth of field","image":"https://example.com/first-frame.jpg","duration":5,"resolution":"540p","generateAudio":true},"webhookUrl":"https://example.com/webhooks/pollo"}
Schema
Input fields for the selected endpoint and mode.
Field
Type
Required
Description
prompt
string
No
Text prompt describing the content to generate.
image
string
Yes
First-frame / reference image URL (HTTP(S).
duration
integer
No
The duration of the generated video, in seconds.
resolution
string
No
Output resolution of the generated media (e.g. 720p, 1080p, 2K, 4K). Values are case-insensitive.
generateAudio
boolean
No
Generate natural-sounding, fitting audio for the output video.
Vidu Q4 combines visual references, recognizable voices, and coordinated shot changes to create character-driven videos. Available as Vidu Q4 Preview, it supports image-to-video and reference-to-video generation.
Run Vidu Q4 on Pollo API to add reference-guided scene creation to storytelling apps, character animation tools, and creative workflows through a shared generation interface.
Key Features of Vidu Q4 API
Multi-Image Scene Guidance
Combine character, environment, and style references to establish a scene’s visual identity, helping recurring subjects remain recognizable as their actions and camera angles change.
Voice Reference Conditioning
Use voice samples to guide a character’s vocal identity, pairing recognizable speech with expressive delivery and synchronized lip movements in reference-driven scenes.
Coherent Shot Switching
Create sequences that move between camera positions while maintaining subject consistency, giving short narratives room for an establishing view, an interaction, and a reaction.
Synchronized Speech and Sound
Generate audio alongside video, including dialogue and sound effects, so character speech and audible actions can develop together within the same scene.
Fluid Action and Effects
Animate running, martial arts, and other active movements while incorporating smoke, particles, or energy effects, making illustrated action scenes useful starting points for animation.
Use Cases of Vidu Q4 API
Recurring Character Episodes: Reuse character and setting references to develop short scenes around a recognizable cast, with voice samples guiding their vocal identities.
Illustrated Dialogue Scenes: Turn character artwork into exchanges with spoken lines, facial reactions, and camera changes that follow the conversation.
Game Character Previsualization: Combine character designs, environment art, and voice references to explore interactions before committing to a complete animation pipeline.
Mascot-Led Brand Stories: Place an established mascot in referenced settings and create short spoken scenarios around its personality and appearance.
Action Storyboard Animation: Animate a prepared frame into a chase, training sequence, or effects-driven encounter with clearly directed movement.
Language-Learning Scenarios: Create short visual conversations with distinct speakers, concise dialogue, and contextual gestures for practice materials.
Reference-Based Creative Tools: Build interfaces where users assemble character, location, and voice assets before directing a scene through a written brief.
How to Use Vidu Q4 API
Get an API Key: Create a Pollo API account and generate your API key from the developer dashboard.
Choose the Model: Send requests to the Vidu Q4 endpoint at /generation/vidu/viduq4/video.
Add Your Inputs: Provide an image for image-to-video, or image/audio items in refs with a prompt for reference-to-video. Set generateAudio as needed.
Generate and Retrieve: Submit the task, poll the returned task status, and download the finished video once it succeeds.
Prompting Best Practices for Vidu Q4 API
Let references establish appearance and vocal identity. Use the prompt to explain who acts, who speaks, how the camera changes, and what carries across each shot.
Give Each Reference a Role: Explain which image defines each character, which establishes the location, and which supplies the intended visual style.
Keep Speaker Assignments Clear: Name the speaker before each line and connect voice references to the intended character without ambiguous pronouns.
Write Shots in Order: Describe the opening view, the important action, and the reaction shot in sequence so the narrative has a clear progression.
Preserve Specific Details: Identify the clothing, props, or spatial relationships that should persist when the camera changes position.
Leave Room for Performance: Keep dialogue brief enough for pauses, gestures, and reactions. Avoid compressing several exchanges and major actions into one short scene.
Example Prompts
Two-Character Dialogue
“Use the supplied baker, courier, and bakery references. Open on both characters beside the counter. The courier asks, ‘Did I miss breakfast?’ Cut closer as the baker slides over a warm roll and replies, ‘Only just.’ Match each character to their supplied voice reference.”
Game Scene Previsualization
“Use the mechanic and underground workshop references. Begin over her shoulder as she tightens a brass valve. Steam escapes; she steps back and laughs. Cut to her relieved expression as she says, ‘That should hold.’ Preserve her patched sleeves, goggles, and the machine’s circular pressure gauge.”
Illustrated Action Animation
“Animate the supplied ink illustration of a mountain ranger crossing a broken bridge. She tests one plank, runs three steps, then jumps onto the far ledge. Track sideways with her movement. Keep the brushwork and red scarf consistent. Add creaking wood, rushing water, and a sharp landing.”
Mascot Storytelling
“Use the supplied otter mascot and greenhouse references. Open wide as the otter discovers a drooping seedling. Move closer while it waters the pot, then cut to its hopeful face. In the referenced voice, it whispers, ‘Your turn to grow.’ Include gentle water sounds and quiet birdsong.”
Vidu Q4 vs Vidu Q3 Pro vs Pollo 2.5
Capability
Vidu Q4
Vidu Q3 Pro
Pollo 2.5
Workflows on Pollo API
Image-to-video and reference-to-video
Text-to-video and image-to-video
Text-to-video and image-to-video
Visual starting point
Starting image or multiple visual references
Written scene or starting image
Written scene or starting image
Voice reference input
Supported in reference-to-video
Not listed on the model page
Not listed on the model page
Generated audio
✅
✅
✅
Shot direction
Intelligent camera switching with reference-guided consistency
Intelligent scene switching and character animation
Multi-shot sequences with style, motion, and camera direction
Recommended For
Scenes built around supplied characters, settings, and voices
Text- or image-led storytelling within the Vidu family
Prompt-led multi-shot creative workflows
Why Choose Vidu Q4 API?
Choose Vidu Q4 when a scene starts with an existing cast, setting, and voice. Its reference guidance helps connect those assets through performance and shot changes.
Pollo API provides access to Vidu Q4 alongside 300+ image and video models, with shared authentication, documented task handling, and human support.
Try Vidu Q4 on Pollo API with a character reference and a short exchange to explore visual continuity, voice guidance, and expressive reactions.
Vidu Q4 API FAQs
What is Vidu Q4?
Vidu Q4 is Vidu’s video generation model, available as Vidu Q4 Preview on Pollo API. It combines image and voice references with synchronized audio, character performance, and coordinated camera changes.
How do Vidu Q4’s two generation workflows differ?
Image-to-video animates a supplied starting image. Reference-to-video uses multiple visual references and optional audio references to guide subjects, settings, style, and voices, making it useful for assembling a scene.
Can Vidu Q4 generate videos from text alone?
The current Pollo API page lists image-to-video and reference-to-video. Supply a visual input and use text to direct the action, dialogue, and camera work rather than relying on text alone.
Can I guide character voices in Vidu Q4?
Yes. Reference-to-video accepts audio references to guide vocal identity. Assign each sample clearly to its intended speaker and keep dialogue concise so speech, facial movement, and reactions have room.
When should I choose Vidu Q4 over Vidu Q3 Pro?
Choose Vidu Q4 when multiple visual references and voice guidance are central to your scene. Vidu Q3 Pro remains an option for text- or image-led generation with synchronized audio.
