Visionreward Video
VisionReward-Video is a chat model developed by Z.ai for aligning video generation with human preferences. It is best at video quality assessment and preference prediction, surpassing VideoScore by 17.2%.
Input
Output
Context
2K
Max Output
-
Parameters
12.5B
Technical Specifications
Model TypeChat
Context Window2,048 tokens
Max Output TokensNot available
Parameters12.5B
Release DateDec 10, 2024
Training CutoffDec 30, 2024
Licenseother
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Estimated GPU Requirements for Visionreward Video
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
More Models by Z.ai
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jul 30, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.