curl-X POST 'https://pollo.ai/api/platform/v1/generation/vidu/vidu-v2/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", "imageTail": "https://example.com/end-frame.jpg", "duration": 4, "resolution": "720p", "movementAmplitude": "auto" }}'
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","imageTail":"https://example.com/end-frame.jpg","duration":4,"resolution":"720p","movementAmplitude":"auto"},"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).
imageTail
string
No
Tail (end-frame) image URL
duration
number
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.
Vidu 2.0 uses a reference image to keep characters, objects, and settings consistent from the first frame to the last as they move.
With Vidu 2.0 on Pollo API, developers can turn a reference image plus a text prompt into a short animated clip, choosing 720P or 1080P output and 4 or 8 second lengths.
Key Features of Vidu 2.0 API
Reference-Anchored Consistency
Start from an image to preserve the subject's face, shape, and surroundings across the whole clip, so a character does not drift into a different look mid-motion.
Start-and-End Frame Control
Supply a first image and a tail image to animate a smooth path between them, giving predictable transitions instead of guessing where the motion lands.
Adjustable Movement Amplitude
Dial motion from subtle to pronounced so a gentle sway and a sweeping action can both come from the same clip, matching the energy each shot needs.
Text-Directed Animation
Use a written prompt to direct the referenced subject’s actions, gestures, and pacing while the image keeps identity and staging consistent.
Coherent Short Clips
Generate four- and eight-second clips that stay stable frame to frame, with natural animation that avoids flicker and warping in single-subject shots.
Use Cases of Vidu 2.0 API
Storybook Character Animation: Bring an illustrated character to life across a scene while keeping its drawn style, colors, and proportions unchanged throughout the clip.
Brand Mascot Motion: Animate a fixed mascot or logo character so marketing tools can generate on-brand moving clips without redrawing the character each time.
Product-in-Hand Demos: Animate a specific product from a reference photo so the item stays visually identical while a hand rotates, opens, or presents it.
Guided Transitions: Use a start image and a tail image to script a clean before-and-after motion, like a closed to open or empty to full change.
Concept Art to Movement: Turn static character or environment art into short moving shots for previz, game pitches, and animatic tools.
Recurring Character Series: Reuse the same reference across many prompts so an app can produce a consistent cast over an ongoing set of clips.
Avatar and Portrait Motion: Animate a portrait into a short expressive clip while preserving the person's features and framing from the source image.
How to Use Vidu 2.0 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 2.0 endpoint at /generation/vidu/vidu-v2-0.
Add Your Inputs: Provide an image URL as the anchor, add a prompt for the motion, then set resolution (720p or 1080p), length (4 or 8), and movementAmplitude.
Generate and Retrieve: Submit the task, poll the returned task status, and download the finished video once processing succeeds.
Prompting Best Practices for Vidu 2.0 API
Because the reference image carries identity, spend your prompt on motion. Describe what the subject does, how strongly it moves, and the mood, so the animation matches the still.
A Simple Prompt Formula
Referenced subject + specific action + movement intensity + camera behavior + setting and mood
What You Should Notice
Let The Image Set Identity: Do not re-describe the subject's appearance in detail; the reference already holds its look, so focus words on action.
Name The Motion Clearly: State the exact movement, such as turns, lifts, or steps forward, so the animation has a definite target.
Match Movement Amplitude: Choose subtle motion for calm shots and larger amplitude for dynamic action to avoid over- or under-animated results.
Plan Tail Frames: When using a start and end image, keep both compositions close so the model can bridge them smoothly.
Keep One Action Per Clip: Focus each 4 or 8 second clip on a single movement so the subject stays coherent throughout.
Example Prompts
Illustrated Character Walk
"The illustrated fox in the reference walks along a mossy forest path, tail swaying gently, ears twitching at a sound. Soft dappled light, medium movement amplitude, slow camera pan following alongside, storybook style preserved from the source image."
Mascot Wave
"The clay mascot from the image lifts one arm and waves cheerfully, then gives a small bounce on its feet. Bright flat studio background, small to medium movement, static camera framing the mascot head to toe."
Guided Transition
"Using the start and end frames, the closed origami crane slowly unfolds its paper wings and lifts into the air. Calm indoor light, medium movement amplitude, gentle upward camera tilt tracking the rising crane."
Product Presentation
"The ceramic teapot from the reference stays centered as a hand tilts it to pour a thin stream into a cup below. Warm window light, small movement amplitude, close side-on camera, clean realistic look matching the source."
Vidu 2.0 vs Kling 2.1 vs Pika 2.2
Capability
Vidu 2.0
Kling 2.1
Pika 2.2
Reference-image consistency
✅ Strong subject and scene lock
✅ Good, less reference-focused
✅ Decent identity hold
Start-and-end frame control
✅
✅
✅
Movement intensity control
✅ Adjustable amplitude
✅ Motion cues via prompt
❌
Resolution options
720P and 1080P
Up to 1080P
Up to 1080P
Clip length
4 or 8 seconds
Short directed clips
Short clips
Recommended For
Consistent character and object animation
Fluid general motion
Quick stylized edits
Why Choose Vidu 2.0 API?
Vidu 2.0 fits products that need the same character, mascot, or product to move without losing its identity, backed by tail-frame control and adjustable motion.
Through Pollo API, you reach Vidu 2.0 with one API key alongside 300+ leading video and image models, with clean docs, task status polling, and generation.
Vidu 2.0 API FAQs
What is Vidu 2.0?
Vidu 2.0 is a Vidu video generation model that animates a reference image with a text prompt, keeping characters, objects, and environments consistent across smooth, natural motion.
Does Vidu 2.0 API need a reference image?
Yes. Generation is anchored to an image URL, with the prompt directing motion, which is why the model holds subject identity so reliably throughout each clip.
What resolutions and lengths does it support?
Vidu 2.0 outputs 720P or 1080P video, with clip lengths of 4 or 8 seconds chosen per request to fit different quality and pacing needs.
Can I control the transition between two images?
Yes. By adding a tail image, Vidu 2.0 animates a path from your start frame to the end frame, useful for scripted before-and-after motion.
How do I control how much the subject moves?
Movement amplitude can be set from small to large, letting you keep motion subtle for calm shots or pronounced for more dynamic action.
Why run Vidu 2.0 through Pollo API?
Pollo API gives one integration for Vidu 2.0 and 300+ other models, with documentation, task tracking, and lower-cost generation.
