Confidential AI

TEE-Terminated TLS

Confidential inference on Intel TDX with per-workload attestation receipts

Confidential AI (Inexorable, Inc.) serves open-weight models inside Intel TDX confidential VMs on its open-source c8s platform, with NVIDIA GPUs in confidential-compute mode. The public /attestation endpoint returns fresh nonce-bound c8s receipts for every workload in the serving cluster — gateway, router, inference workers, and supporting services — including Intel TDX quotes, pinned image measurements, mesh identity proofs, and NVIDIA GPU evidence. A post-quantum over-encryption tunnel (ML-KEM-768 + X25519) terminates inside the attested TEE.

Models2

From$0.20/M input tokens

Organization rolesStorefront · Workload operator · TEE platform

Advertised ChannelTEE-Terminated TLS

Post-Quantum KEXPQ hybrid to attested TEEX25519MLKEM768 · measured 2026-09-21

API Features

OpenAI Compatible✓

Streaming✓

Function Calling—

Batch Inference—

Vision—

Embeddings—

Attestation Status

View full details →

Last checked: Sep 21, 2026, 5:01 PM

Model coverage: 0/2 current catalog models have an explicit check.

Incomplete evidenceHardware and TCB passed, but one or more later evidence gates are not proven.

0 verified1 partial0 failed0 unavailable

Hardware1/1 goodTCB policy1/1 goodFreshness1/1 goodChannel binding1/1 goodWorkload identity1 warningSource provenance1 warningRequest route1 warningAvailability1/1 good

Workload identity not proven 1Source provenance not proven 1Request route not proven 1

◐Confidential AI cluster (c8s)Intel TDXTEE-Terminated TLSDetails

Get notified when this provider's attestation verdict changes:Alerts RSS

Models & Pricing

Model	Family	Context	Max Out	Input /M	Output /M	Free	Attestation	Observed route
DeepSeek V4 Flash	DeepSeek	1024K	—	$0.20	$0.40	—	No current check	Not identified
MiniMax M3 (MXFP8)	MiniMax	1024K	—	Bespoke deployment — no public per-token price; contact hello@confidential.ai.	—	—	No current check	Not identified