Picsart-AI-Research/Text2Video-Zero

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

4245 stars · 388 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Official implementation of Text2Video-Zero, a zero-shot text-to-video generation method that adapts text-to-image diffusion models like Stable Diffusion to video without any training. It supports text-to-video synthesis, pose/edge/depth-conditioned generation, and instruction-guided video editing (Video Instruct-Pix2Pix).

Use cases

• generate videos from a text prompt
• zero-shot text-to-video generation with stable diffusion
• edit videos with text instructions
• generate videos conditioned on pose or edge maps
• create temporally consistent videos without training a video model
• run text-to-video on a consumer GPU with limited VRAM

When to choose

• you want text-to-video generation without training on video datasets
• you want to reuse existing Stable Diffusion or DreamBooth models for video
• you need instruction-guided video editing
• you have a GPU with ~7-12 GB VRAM and want low-cost video generation

When to avoid

• you need production-grade, actively maintained software
• you require the highest-quality modern text-to-video output
• you have no GPU available
• you need a permissive license - the license is non-standard

Facets

library · maturity maintenance

video-processing machine-learning deep-learning image-processing deep-learning artificial-intelligence computer-vision python text-to-video stable-diffusion diffusion-models zero-shot video-editing instruct-pix2pix research-code iccv-2023 video gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Picsart-AI-Research/Text2Video-Zero")