AI Forensic Engine — multi-detector ensemble for AI-generated text and image authentication.
Project description
███████╗██╗ ██╗███╗ ██╗████████╗██╗ ██╗
██╔════╝╚██╗ ██╔╝████╗ ██║╚══██╔══╝██║ ██║
███████╗ ╚████╔╝ ██╔██╗ ██║ ██║ ███████║
╚════██║ ╚██╔╝ ██║╚██╗██║ ██║ ██╔══██║
███████║ ██║ ██║ ╚████║ ██║ ██║ ██║
╚══════╝ ╚═╝ ╚═╝ ╚═══╝ ╚═╝ ╚═╝ ╚═╝
Installation • Quick Start • CLI Reference • Configuration
Catch AI fakes in seconds.
Synth is a tool that tells you if an image or document was created by an AI or a human.
What is Synth?
Synth acts like a digital detective. Whenever you encounter a picture, a PDF, or text and wonder, "Did a computer make this?", Synth gives you the answer.
Instead of relying on just one check, Synth uses a team of specialized AI models. Each model looks at the file from a different angle, and they all vote to give you a highly accurate, final verdict. Everything runs safely and privately on your own computer.
Features
- 🔍 Reads Anything — Just point it at an image or a PDF. It extracts text and analyzes images automatically.
- 🤖 Team of Experts — Uses an ensemble of state-of-the-art models that vote together to catch fakes.
- ⚡ Adjustable Depth — Choose between a quick glance or a deep, forensic investigation.
- 🍎 Works Anywhere — Runs smartly on Mac, Windows, and Linux with automatic hardware acceleration.
- 🔒 100% Private — No cloud uploads needed. Your files stay perfectly private.
Installation
Getting started is simple. You can install Synth directly from PyPI.
[!NOTE] Since the name
synthwas already taken on PyPI, the package is published under the namesynth-authenticator.
Run the following command to install using pip:
pip install synth-authenticator
Alternatively, you can install it from source:
git clone https://github.com/khushalv21/SYNTH.git
cd SYNTH
pip install .
Note: Synth will automatically detect your system's hardware to run as fast as possible. The first time you run it, it will download the necessary AI models.
Quick Start
Synth is designed to be effortless. Just type synth followed by the file you want to check:
synth my_picture.png
synth my_document.pdf
Choose your investigation depth
You can tell Synth how closely you want it to look using "Profiles":
synth my_picture.png --profile fast # A quick check
synth my_picture.png --profile balanced # The normal check (recommended)
synth my_picture.png --profile forensic # A deep, thorough investigation
Scan a whole folder
Have a lot of files? You can scan a whole directory at once:
synth ./my_documents_folder/
CLI Reference
| Command | Description |
|---|---|
synth <file> |
Analyze a file (Image or PDF) |
synth <folder>/ |
Analyze an entire folder |
synth models |
See all the internal AI models doing the work |
synth help |
Show the help menu |
Tip: If you only want to extract the text without running an AI check, you can use the --no-text flag to toggle the text panel.
Configuration & Advanced Use
For developers or those needing advanced setup (like plugging in external APIs such as OpenAI or Anthropic), simply create a .env file in the directory. You can use the provided .env.example as a template.
For full technical documentation, see the Architecture Guide and API Guide.
License
MIT — see LICENSE for details.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file synth_authenticator-2.0.0.tar.gz.
File metadata
- Download URL: synth_authenticator-2.0.0.tar.gz
- Upload date:
- Size: 63.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3dd0c3af282b0cf6f9c89a4367159815922428819288587274918d2594ba0b1a
|
|
| MD5 |
7f8c131951d331e6474cc46f3a54aa1f
|
|
| BLAKE2b-256 |
5f52901d60a2dabc94bfb6ac3693281743620a1f3277319cbec62d0a1038cc7a
|
File details
Details for the file synth_authenticator-2.0.0-py3-none-any.whl.
File metadata
- Download URL: synth_authenticator-2.0.0-py3-none-any.whl
- Upload date:
- Size: 77.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b343c383b2f7e2d5dc0e67a8f9682d77dc35015938ebb4c33c00d74e1fb749b1
|
|
| MD5 |
f6215b33628612a96194a34e99572903
|
|
| BLAKE2b-256 |
8e973a531e135eca5904b26a81cfe989254ea1c2d76e55e21ddeb55f58f2378c
|