Live Confidential AI Prices
Read from the same live pricing feed as /pricing: Confidential VMs per hour, TEE models per million tokens. One competitor price, sourced and dated.
Confidential Compute
Intel TDXEvery Confidential VM tier in the feed right now, each an Intel TDX trust domain. Price of the whole VM per hour, and per GPU-hour on 8-GPU nodes.
RTX PRO 6000 Blackwell
Intel + NVIDIAH100
Intel + NVIDIA8x H100
Intel + NVIDIAH200
Intel + NVIDIA8x H200
Intel + NVIDIA8x B200
Intel TDX quoteConfidential GPU, Price Comparison
Azure's NCC40ads H100 v5 (AMD SEV-SNP with one H100 NVL), the only hyperscaler confidential H100 with a public on-demand price (Google prices its A3 High confidential H100 only on Spot and Flex-start, and its G4 RTX PRO 6000 on demand), next to our single-GPU H100 Confidential VM (Intel TDX).
H100, single GPU
Confidential VM vs NCC40ads H100 v5Google Cloud sells confidential H100 on A3 High at per-zone rates and AWS has no confidential GPU. Dated, sourced comparison: confidential GPU providers compared.
Confidential AI Inference, Price per Million Tokens
Every TEE model in the feed, priced per million input and output tokens. The model runs inside Intel TDX at our inference provider; our API relays requests in clear.
Qwen3.8-27B-TEE
gemma-4-31B-turbo-TEE
Kimi-K2.6-TEE
Qwen3.5-27B-TEE
DeepSeek-V3.2-TEE
GLM-5.1-TEE
Qwen3.6-27B-TEE
GLM-5.2-TEE
Kimi-K3-TEE
DeepSeek-V4-Flash-0731-TEE
Why VoltageGPU
Per-second billing
Only pay for what you use. Stop your pod and billing stops instantly. No rounding to the hour.
No lock-in
Stop anytime, no commitment, no reserved instances. Prepaid balance: a VM's first hour is charged at start and the unused part is refunded when you release it.
11 TEE Models
Confidential AI inference via OpenAI-compatible API. Every model runs inside an Intel TDX hardware enclave.
Start building today
$3.80 welcome credit with your first top-up of $10 or more. Per-second billing, no subscription.