AI Trends

FastVideo Open-Sources Dreamverse Harness, an Agent Framework for Video Production

UCSD Hao AI LabUpdated Nov 20 04:48

AI summary

FastVideo has open-sourced Dreamverse Harness, enabling AI agents to assist with the video production process from scripting and storyboarding through final editing, with support for fully local operation or cloud services. The framework can schedule models and shot-generation tasks across up to four NVIDIA DGX Spark systems, supports conversational scene editing, and can upscale selected shots to 2K. Built on DeepSeek Harness, it allows models to be swapped and workflow steps to be added or removed, while support for Blender and ComfyUI is still under development.

Why read this

SelectedAI score 76

Its complete workflow—from scripting and storyboarding to editing and upscaling—along with its local multi-machine scheduling design provides a reusable solution for agent-based video production.

Topics

Tags

Body

FastVideo is open-sourcing Dreamverse Harness 🚀

Since we released FastH3, the community has picked it up widely, especially on local machines. But a polished film takes more than a single clip.

FastVideo Dreamverse Harness enables any agent to work with you through the whole video production, from the first script to the final cut. And you can run all of it fully local on your own NVIDIA DGX Sparks or through cloud providers.

📝 Blog: https://t.co/Uk3RoE5rUC 💻 GitHub: https://t.co/Xcvadrksnd

@NVIDIARTXSpark @NVIDIAAI @MiniMax_AI

#OpenSourceHiggsfield

Source

UCSD Hao AI Lab