jiachenli-ucsb/T2V-Turbo-MS

T2V-Turbo: Breaking the Quality Bottleneck of Video Consistency Model with Mixed Reward Feedback

Model description 🚀

This repository contains unet_lora.pt that can turn ModelScopeT2V into our T2V-Turbo (MS). Our T2V-Turbo (MS) can achieve both fast and high-quality T2V generation. On VBench, the 4-step generation from our T2V-Turbo (MS) even outperform proprietary systems, including Gen-2 and Pika. Please refer to our GitHub repo for detailed instructions.

4-step Text-to-video Generation

a dog wearing vr goggles on a boat	Mickey Mouse is dancing on white background	close-up shot, high detailed, A boy with a baseball cap, freckles, and a playful grin.

an old man with a long grey beard and green eyes, camera rotate anticlockwise	The flowing water sparkled under the golden sunrise in a peaceful mountain river.	close-up shot, high detailed, a girl with long curly blonde hair and sunglasses

Misuse, Malicious Use and Excessive Use 📖

Our model is meant for research purposes.

• It is prohibited to generate content that is demeaning or harmful to people or their environment, culture, religion, etc.
• Prohibited for pornographic, violent and bloody content generation.
• Prohibited for error and false information generation.