happyhorse-1.0-i2v
HappyHorse-1.0-I2V is a Tusheng video variant that inputs an image and text prompt and outputs smooth and natural high-quality video. Its core capability lies in "highly restored dynamic picture generation", which can accurately understand text semantics, combine visual information of static images with text descriptions, and generate video clips with rich details and natural movements. The input modalities of this model are images and text, and the output is video. It provides inference services through the Alibaba Cloud Bailian platform and is deployed in Beijing, Singapore, Frankfurt, Virginia and other regions.
- Context
- 0 tokens
- Max output
- 0 tokens
- Input
- llm image video
- Output
- video
Pricing
Price: $0.24 / request
720P image_to_video : $0.14 /
1080P image_to_video : $0.24 /
Endpoints
- openai