# SCLBD/DeepfakeBench

A comprehensive benchmark of deepfake detection

Repository: https://github.com/SCLBD/DeepfakeBench
Canonical: https://ross.abutalabs.com/products/deepfakebench
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-08-20T08:03:47+00:00

## Health v2 (maintenance only)
Score: 36/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 37, release rhythm 8, longevity 84
- inputs: {"age_days": 1184, "days_push": 378, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1093, forks 183 (observed 2026-08-28T04:03:33.724705+00:00)

## What it is
DeepfakeBench is a comprehensive benchmark framework for deepfake detection, providing a unified platform for data management, implementation of state-of-the-art detection methods, and standardized evaluation protocols. It was published at NeurIPS 2023 (Datasets & Benchmarks) and includes pre-trained weights and integration with datasets like Celeb-DF, FF++, and DF40.

## Use cases
- benchmark deepfake detection methods under a unified evaluation protocol
- train and evaluate face forgery detection models on standard datasets
- compare state-of-the-art deepfake detectors reproducibly
- test detection models on new deepfake datasets like DF40
- build new deepfake detection methods on an integrated framework
- evaluate detectors on both face deepfakes and synthetic images

## When to choose
- you need standardized, reproducible comparison of deepfake detection methods
- you want pre-trained weights and unified data pipelines for face forgery detection research
- you are developing a new detector and want a common training/evaluation framework

## When to avoid
- you need a production-ready deepfake detection service rather than a research benchmark
- your use case requires a permissive commercial license (project is CC BY-NC 4.0)
- you work outside Python/PyTorch environments

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, benchmarking, image-processing, computer-vision, testing
- domain: deep-learning, computer-vision, artificial-intelligence, security
- platform: python
- tags: deepfake-detection, benchmark, pytorch, face-forgery, media-forensics, reproducibility, research, linux, gpu

## Member repositories
- SCLBD/DeepfakeBench (main) score 36

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:33.724705+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 2026-08-30T06:47:40.430317+00:00, confidence not recorded.
  - readme: https://github.com/SCLBD/DeepfakeBench (fetched 2026-08-28T04:03:33.724705+00:00, sha 9fdf3d54f71f)
- Data as of 2026-08-30T08:39:29.467469+00:00.