Introducing Content Seal

State-of-the-Art Invisible Watermarking

A state-of-the-art framework for invisible, robust watermarking across all modalities, audio, image, video, and text, with open-source implementations.

Read the Research Papers Get the Models
Research Overview

Watermarking across all modalities

Content Seal offers state-of-the-art proprietary and open-source tools for embedding robust and imperceptible watermarks across images, video, audio, text, and generative AI models. As AI-generated content becomes increasingly sophisticated, provenance tracking and authentication mechanisms are more critical than ever.

Content Seal Image is deployed at scale for Muse Image with a custom proprietary implementation. We also provide open-source versions of our research models for images and video, readily available to download.

Post-hoc watermarking

Watermarks applied after content generation by any model or system. This model-agnostic approach works universally across images, video, audio, and text—no matter how the content was created. Designed for protecting existing content and third-party generated media.

Content Seal for Images and Video

+
Pixel Seal: Adversarial-Only Training for Invisible Image and Video Watermarking

Flagship image & video watermarking model SOTA in terms of robustness and imperceptibility, which is built with a better and more stable adversarial-only training paradigm.

Paper Code
We Can Hide More Bits: The Unused Watermarking Capacity in Theory and in Practice

Bigger model with 4× capacity boost to 1024 bits while preserving quality and robustness.

Paper Code
Video Seal: Open and Efficient Video Watermarking

Extension of image watermarking models to video, resilient to editing and video codecs.

Paper Code Demo
Watermark Anything with Localized Messages

Embed (possibly multiple) localized watermarks into images, survives inpainting and splicing attacks.

Paper Code
Geometric Image Synchronization with Deep Watermarking

Watermarking models for robust image synchronization, enabling to revert geometric transformations applied to image.

Paper Code

Content Seal for Audio

+
Proactive Detection of Voice Cloning with Localized Watermarking

Localized audio watermarking with sample-level detection and streaming support for real-time applications.

Paper Code

Content Seal for Text

+
How Good is Post-Hoc Watermarking With Language Model Rephrasing?

Comprehensive evaluation framework for post-hoc text watermarking with LLM rephrasing.

Paper Code

In-model & generation-time watermarking

Watermarks embedded during the generation process by modifying the model's latent representations or embedding the watermark directly in the model weights. This native approach enables seamless integration with diffusion models, LLMs, and other generative systems for real-time watermarking as content is created.

Content Seal for Text

+
TextSeal: A Localized LLM Watermark for Provenance & Distillation Protection

State-of-the-art LLM watermark with dual-key Gumbel-max sampling, entropy-weighted scoring, and multi-region localization. Distortion-free, preserves reasoning performance, robust to dilution, and radioactive through distillation.

Paper Code

Content Seal for Image & Audio

+
Learning to Watermark in the Latent Space of Generative Models

Unified latent space watermarking that enables 20× speedup over pixel methods and secures open-source models via in-model distillation.

Paper Code
The Stable Signature: Rooting Watermarks in Latent Diffusion Models

Roots the watermark in the model's latent decoder for tracing the outputs of latent generative models.

Paper Code
Watermarking Autoregressive Image Generation

Watermarking for autoregressive image generation models.

Paper Code

Radioactivity

+
Watermarking Makes Language Models Radioactive

Designed to detect if a language model was trained on synthetic text by detecting weak residuals of watermark signals in fine-tuned LLMs, with high confidence detection even when as little as 5% of training text is watermarked.

Paper Code
Detecting Benchmark Contamination Through Watermarking

Watermarks benchmarks before release to detect if models were trained on test sets, using theoretically grounded statistical tests to identify contamination while preserving benchmark utility.

Paper Code

Watermark security

Research on adversarial attacks and defenses for watermarking systems. Red teaming efforts explore vulnerabilities like watermark removal, forgery, and spoofing attacks—while developing robust countermeasures to help watermarks remain secure and reliable.

Transferable Black-Box One-Shot Forging of Watermarks via Image Preference Models

Black-box watermark forging using image preference models for red-teaming watermarking systems.

Paper Code
Try it yourself

Get the Models

Access the open-source versions of our research models. Open weights, MIT Licensed, and designed for immediate scientific exploration.

Content Seal for Video & Image Content Seal for Audio Content Seal for Text
watermark_detect.py
Applications

Learn More

Discover how invisible watermarking technology is being explored to help address content provenance challenges across media platforms.