# Netflix/void-model

Repository: https://github.com/Netflix/void-model
Canonical: https://ross.abutalabs.com/products/void-model
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-06-20T11:14:30+00:00

## Health v2 (maintenance only)
Score: 54/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 88, release rhythm 35, longevity 11
- inputs: {"age_days": 160, "days_push": 74, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1965, forks 185 (observed 2026-08-28T04:05:59.817788+00:00)

## What it is
VOID (Video Object and Interaction Deletion) is a research model from Netflix that removes objects from videos along with the physical interactions they induce on the scene, built on CogVideoX and fine-tuned for interaction-aware video inpainting. It ships transformer checkpoints via Hugging Face with a Gradio demo and Colab notebook.

## Use cases
- remove a person from a video and make held objects fall naturally
- erase objects from video clips including shadows and reflections
- interaction-aware video inpainting with masks
- research on video object removal models
- run video inpainting with CogVideoX checkpoints

## When to choose
- you need object removal from videos that accounts for physical interactions
- you want a state-of-the-art video inpainting model with pretrained checkpoints
- you are doing computer vision research on video editing

## When to avoid
- you need a production-ready video editing pipeline with Diffusers support (still a TODO)
- you lack GPU resources for large video diffusion inference
- you only need simple 2D image inpainting

## Facets
- artifact type: library
- maturity: active
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: computer-vision, machine-learning, artificial-intelligence
- platform: python
- tags: video-inpainting, object-removal, cogvideox, diffusion-model, eccv-2026, research, video, gpu

## Member repositories
- Netflix/void-model (main) score 54

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:59.817788+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:05:27.953375+00:00, confidence not recorded.
  - readme: https://github.com/Netflix/void-model (fetched 2026-08-28T04:05:59.817788+00:00, sha bcd07b0f5c35)
- Data as of 2026-08-30T08:39:29.467469+00:00.