# facebookresearch/VMZ

VMZ: Model Zoo for Video Modeling

Repository: https://github.com/facebookresearch/VMZ
Canonical: https://ross.abutalabs.com/products/vmz
Language: Python
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2025-06-17T00:43:51+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 27, release rhythm 35, longevity 100
- inputs: {"age_days": 3052, "days_push": 443, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1052, forks 158 (observed 2026-08-28T04:03:23.556290+00:00)

## What it is
VMZ is a model zoo from Facebook AI's Computer Vision team providing Caffe2 and PyTorch implementations of video classification models such as R(2+1)D, MCx, CSN, and IG-65M pre-trained models. The repository has been archived and is no longer supported.

## Use cases
- reproduce video action recognition research results
- use pretrained video models for video classification
- fine-tune R(2+1)D or CSN models on my own video dataset
- leverage IG-65M pretrained weights for video understanding
- train audio-visual models with gradient blending

## When to choose
- you need strong pretrained video action recognition models like R(2+1)D or CSN
- you want to reproduce Facebook AI's video modeling research results
- you can work with Caffe2 or older PyTorch codebases

## When to avoid
- you need actively maintained code or support
- you want modern video models or recent PyTorch compatibility
- you need a reliable PyTorch CSN implementation (noted as buggy)

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, computer-vision, video-processing
- domain: computer-vision, machine-learning, deep-learning
- platform: python
- tags: video-modeling, action-recognition, model-zoo, pretrained-models, caffe2, pytorch, archived, video, linux, gpu

## Member repositories
- facebookresearch/VMZ (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:23.556290+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 2026-08-30T06:59:43.046796+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/VMZ (fetched 2026-08-28T04:03:23.556290+00:00, sha 4b9be2481557)
- Data as of 2026-08-30T08:39:29.467469+00:00.