image-to-video
Video
Audio
Ultra fast and controllable avatar generation.
Performance
High quality video generation in less than 15s.
Multi-modal
Image, audio, and text input.
Up to 3 minute video.
Aesthetic
Control
Cinematic-quality avatars with precise performance control.
P-Video-Avatar is the most efficient model for high quality avatar generation.
Acknowledgments
The launch of P-Video-Avatar is a milestone for real-time, high-quality video avatar generation.
But it is not a journey we took alone. We want to acknowledge the teams whose work defined the state-of-the-art for visual generation, introduced the proper techniques, and set the high bar for quality that we aimed to match including
Black Forest Labs for their work on the Flux series,
Alibaba for their work on Qwen Image, Qwen Image Edit, Z-Image, and Wan Video,
Lightricks for their work on LTX,
Hugging Face for their work on Diffusers and Transformers,
the Open source community the fantastic job on countless open models, open packages, and open research.
Pruna is a member of the Coalition for Sustainable AI.
Pruna co-authored "Key challenges for the environmental performance of AI".
Code Carbon is integrated into Pruna.