Secure AI inference, backed by hardware and proof
Every model in the Enclave42 catalogue runs as a confidential workload, processed inside an Intel TDX trust domain on an NVIDIA GPU. You don’t have to trust that claim — you can verify it yourself.
Confidential by hardware, not by policy
When you send a request, it is processed inside an Intel TDX trust domain on an NVIDIA GPU. That’s a hardware boundary, not a policy promise.
Customer-verifiable, not vendor-asserted
Requesting a bundle produces the raw TDX quote and NVIDIA GPU evidence, a verification manifest bound to a nonce you generate yourself, and a short verification script you can read before running. Verification runs offline, on your own machine, against Intel and NVIDIA — not against Enclave42.
What this protects
It proves that a specific runtime instance was executing inside an Intel TDX trust domain on an attested NVIDIA GPU at the moment your nonce was issued, and that the encryption key offered to you belongs to that instance.
What this does not protect
It does not prove anything about instances you have not attested, and it does not replace your own review of what the workload does with your data once decrypted inside the enclave. Attestation establishes the boundary — it does not audit the code inside it. It also does not, by itself, establish where the workload is physically located.