Deepfake detectionComputer vision

TrustMark

An image watermark for arbitrary resolutions built for the Content Authenticity Initiative: it can both apply a mark and remove one. The detector errs in both directions - a human reviews the output.

Developer
Adobe Research and University of Surrey, USA
First release
Nov 2023
Latest release
Sep 2026
Sizes
model types Q and P with different mark capacity
License
Commercial use allowedMIT, including the downloadable model files
Running
On your own serverAlso runs without a GPU
Industries
Media and production, Marketing and content, Security, Legal
Deploy on your server GitHub

What it does

  • Marking images on the way out of your own pipeline
  • Checking the provenance of a submitted image
  • Linking with content provenance metadata
  • Removing your own mark before publication

Where it is used

Media and photo librariesMarketing and advertisingInformation securityLegal departments

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. Текущие обновления репозитория
  2. Публикация на ICCV 2025
  3. TrustMark (статья и код)

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can TrustMark be used in a commercial project?

Yes. License: MIT, including the downloadable model files. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does TrustMark need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does TrustMark support Russian?

Language does not matter for this model: it does not work with text.

Where can I download TrustMark and what does it cost?

The TrustMark weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Deepfake detectionWatermark Anything (WAM)Meta · USACommercial use allowed

An image watermark that can be applied to individual regions: the model shows which part of the image is marked. It errs in both directions - a human reviews the result.

DetailsDeepfake detectionVideoSealMeta · USACommercial use allowed

A watermark for video and images that survives re-encoding and cropping. The detector errs in both directions: a missing mark does not prove a forgery, and finding one is a reason for a human to check.

DetailsDeepfake detectionAudioSealMeta · USACommercial use allowed

An imperceptible mark in synthetic speech plus a fast detector that finds it even inside a fragment of a long recording. The detector errs in both directions: a hit is a reason for a human to check, not proof.

Details

Source: github.com/adobe/trustmark. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.