showlab/Tune-A-Video

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

4364 stars · 388 forks observed · 2026-08-28

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

Tune-A-Video is the official PyTorch implementation of an ICCV 2023 paper that fine-tunes pre-trained text-to-image diffusion models (like Stable Diffusion) for text-to-video generation using only a single text-video pair. It introduces a spatio-temporal attention mechanism and one-shot tuning strategy, with DDIM inversion for improved consistency.

Use cases

• generate short videos from a text prompt and one example video
• fine-tune stable diffusion for text-to-video generation
• create stylized video variations from a single reference clip
• edit videos with text prompts using diffusion models
• run one-shot video tuning on a GPU
• try text-to-video generation in a colab notebook

When to choose

• you want research-grade one-shot text-to-video generation built on Stable Diffusion
• you have a single reference video and want to generate new videos with different subjects or styles
• you need the reference implementation of the Tune-A-Video paper for academic work

When to avoid

• you need production-quality long or high-resolution video generation
• you lack a capable GPU, since diffusion fine-tuning is compute-intensive
• you want a maintained product with active updates rather than a research codebase

Facets

library · maturity maintenance

machine-learning deep-learning video-processing image-processing artificial-intelligence deep-learning image-processing python text-to-video diffusion-models stable-diffusion one-shot-tuning iccv-2023 research video gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="showlab/Tune-A-Video")