Official

haiper-ai / haiper-video-2

Generate 4s and 6s videos from a prompt or image

  • Public
  • 10K runs
  • $0.05 per second of video
  • Commercial use

Input

*string
Shift + Return to add a new line

Text prompt for video generation

string
Shift + Return to add a new line

Things you don't want in the video

string
Shift + Return to add a new line

A URL to an image to use in the video. Must be a JPEG, PNG, or WebP file.

string

Aspect ratio for the generated video

Default: "16:9"

integer

Number of seconds to generate

Default: 6

boolean

Use prompt enhancer

Default: true

integer

Random seed. Set for reproducible generation

Output

Generated in

Pricing

Official model
Pricing for official models works differently from other models. Instead of being billed by time, you’re billed by input and output, making pricing more predictable.

This model is priced by how many seconds of video are generated.

TypePer unitPer $1
Output
$0.05 / second of video
or
20 seconds of video / $1

For example, generating 100 seconds of video should cost around $5.00.

Check out our docs for more information about how per second of video pricing works on Replicate.

Readme

Haiper 2.0

https://haiper.ai

Haiper 2.0 is a video and image generation model built on perceptual Diffusion Transformer (DiT) architecture. The system generates high-resolution video content with enhanced temporal coherence and reduced generation times compared to previous versions.

Features

Core Capabilities

  • Video and image generation with improved resolution
  • Enhanced temporal coherence for smoother motion
  • Faster generation times
  • Planned upgrade path to 4K resolution

Technical Implementation

  • DiT (Diffusion Transformer) model architecture
  • Neural networks optimized for video generation
  • Advanced temporal consistency algorithms

Feedback Channels