Skip to content
SPECIFICTION
Get listed

Detoxify vs Google Cloud Video Intelligence

Detoxify

github.com

Google Cloud Video Intelligence

cloud.google.com

Side by side

FigureDetoxifyGoogle Cloud Video Intelligence
Price fromFreeFree
Free planYesYes
Free trial—Yes
Open source——
Web—Yes
LinuxYes—
API—Yes
Self-hostedYes—
Plans published12
Spec score · Content Moderation Tools Software7.6 (#12)7.5 (#13)
API accessThe repository says to email [email protected] to request access to its API.Google provides client libraries and a command-line quickstart for interacting with the API programmatically.
PurposeDetoxify provides trained models and code to classify toxic comments from three Jigsaw challenges.The API analyzes stored and streaming video and returns metadata for content discovery and video applications.

Plans

Detoxify

  • Open-source libraryFree

Detoxify pricing →

Google Cloud Video Intelligence

  • Video Intelligence API — stored video annotationFree
  • Video Intelligence API — streaming video annotationFree

Google Cloud Video Intelligence pricing →

In their own words

Detoxify

Detoxify is a free, self-hosted library that provides trained models and code to classify toxic comments. Its original, unbiased, and multilingual models are based on BERT, RoBERTa, and XLM-RoBERTa, respectively. Depending on the model, outputs include categories such as threats, obscenity, insults, identity attacks, and explicit sexual content. The multilingual model is intended for English, French, Spanish, Italian, Portuguese, Turkish, and Russian. The project describes research, fine-tuning on carefully constructed representative datasets, and helping moderators flag harmful content faster as intended uses. It can be installed with pip and called from Python to predict labels for one string or a list. Inference uses CPU by default and accepts a PyTorch device such as CUDA. The project warns that profanity-associated words may be labeled toxic regardless of tone or intent, potentially biasing results against vulnerable minority groups. Inference uses Transformers and PyTorch Lightning; training also requires the Kaggle API to download data. API access can be requested by email. The library is distributed under Apache-2.0.

Visit github.com

Google Cloud Video Intelligence

Google Cloud Video Intelligence is a cloud API that analyzes stored and streaming video and returns metadata for content discovery and video applications. Metadata can be extracted at video, shot, or frame level, and Google says its models recognize over 20,000 objects, places, and actions. Documented functions include label and shot-change detection, explicit-content detection, and face, logo, object, and person detection, as well as speech and text recognition. Streaming annotation can provide near real-time insights and support event triggers based on objects. Google lists content moderation, recommendations, searchable media archives, and contextual ad placement as use cases. Client libraries and a command-line quickstart support programmatic access. Each listed feature has its first 1,000 minutes per month free for both stored and streaming annotation; partial minutes round up. Beyond that amount, rates vary by feature and annotation type. Pricing is listed from $0.05, and a free trial is available. Other Google Cloud resources used by a project may incur separate charges. Celebrity recognition is a restricted-access feature.

Visit cloud.google.com