Back
Blog
Aug 26, 2026

DETECT-World Is Live on Eden AI: Deepfake Detection Through a Single API

CONTENTS
Active heading
Section heading
CONTRIBUTORS
Dev Shah
Head of Developer Relations

Most AI-powered products in 2026 already run through an API gateway. LLMs, OCR, translation, speech processing: developers route all of it through unified platforms so they can swap providers, manage costs, and avoid vendor lock-in.

Detection is the odd one out. It still often gets integrated as a standalone service, with a separate contract, SDK, and billing dashboard, even when everything else on the stack has been consolidated.

DETECT-World, Resemble AI's latest deepfake detection model, is now available on Eden AI as a Day-0 launch integration. If your team already uses Eden AI for other AI capabilities, adding deepfake detection doesn't require a new vendor relationship.

Eden AI is a unified API platform based in France, used by over 200,000 developers. It aggregates 700+ AI models from 50+ providers behind a single endpoint, so teams can access LLMs, computer vision, document parsing, and speech processing through one API key and one billing dashboard.

With DETECT-World now on the platform, deepfake detection plugs into the existing workflow without adding another account or integration to manage.

Detection keeps getting pushed to the backlog

Picture a platform team that already has image classification, OCR, and text analysis running through a unified API. The product manager says, "We need deepfake detection too."

The engineering team looks at what that means: another vendor to evaluate, another SDK to learn, credentials to rotate, and monitoring to set up. It goes on the backlog behind things that are easier to ship, and it stays there.

We hear versions of this from the developer teams we talk to. Teams know deepfake detection matters, but the integration overhead makes it feel like a bigger project than it should be.

Meanwhile, the threat isn't slowing down to wait for anyone's roadmap. Resemble AI's H1 2026 Deepfake Threat Report documented 821 verified attacks in six months with 292.8 billion in media reach, effectively matching the full-year 2025 total in half the time.

That's what this integration is meant to solve. If your team already uses Eden AI, adding DETECT-World follows the same process as adding any other provider on the platform.

Want to test it on your own media? Try Resemble Detect on Eden AI, there's no separate API key needed.

What DETECT-World brings to the platform

Most detection systems recognize the statistical artifacts that specific generators leave behind, which means they can struggle when a brand-new generator launches.

DETECT-World adds a second layer. Alongside artifact scanning, it checks whether lighting, motion, and spatial consistency in a scene actually make sense. That physical-plausibility check helps it detect content from generators it was never trained on.

The model has been benchmarked against more than 250 generative models across 54 languages. On audio, it scored 99.47% accuracy on a third-party Podonos benchmark, ranking first among 18 systems tested. On image and video, internal benchmarks put accuracy at 95.77% and 98.2% respectively.

Read the research on DETECT-World.

Why multi-provider detection matters

Eden AI's architecture lets teams route the same piece of media to multiple detection providers in a single request. When two out of three providers flag content as synthetic, you have a stronger basis for acting on that verdict. And when one provider misses a new generator, another provider can pick it up.

There's a similar idea behind DETECT-World itself. It uses two different detection methods on the same input, so content that gets past one check can still be caught by the other. Eden AI extends that approach across providers.

For teams running identity verification, content moderation, or media intake at scale, the question is what it costs to add detection versus what it costs to miss something.

Through a gateway you already use, the first number is close to zero.

Three endpoints, one API key

Endpoint What it does Input
image/deepfake_detection/resemble Deepfake detection for images Image file
image/ai_detection/resemble AI-generated image detection Image file
video/deepfake_detection_async/resemble Deepfake detection for video (async) Video file


Authenticate with your existing Eden AI key, select Resemble AI as the provider, and submit your media. The response includes a detection verdict and confidence score.

Get started

DETECT-World is live on Eden AI now, and the integration covers image and video.

Access Resemble Detect on Eden AI.

For enterprise deployments, custom detection thresholds, or a walkthrough of how Resemble Detect performs on your specific media types, book a demo today.

Try Resemble AI free
Generate with confidence. Verify ownership. Detect deception. Only with Resemble AI.
Get started
Know what's real — and what's a real threat.
Join thousands of developers and enterprises detecting AI fraud and protecting their content with Resemble AI