Trial ready
NVIDIA H100

Proven confidential inference and fine-tuning capacity.
Memory
80GB HBM3
Bandwidth
3.35 TB/s
Region
US-West
Scale
1-2 GPUs
On-demand
$3.08/GPU/hr
24h minimum
Slot
$2.38/GPU/hr
reserved
GPU TEE Marketplace
H100, H200, and B300 capacity with CVMs, dual attestation, and TEE-aware operations.
Trial a machine for 24 hours, reserve a slot, or quote dedicated clusters. Phala handles the hard part: confidential GPUs, Intel TDX runtime, NVIDIA attestation, and the DevOps required to keep it working.
Confidential GPU cloud



hardware proof rail
GPU TEE
H100
GPU TEE
H200
GPU TEE
B300
Trial
24h minimum
Reserve
slots and clusters
Verify
CVM + GPU evidence
Marketplace inventory
Pick a GPU for a 24-hour trial, reserve a slot for sustained jobs, or quote a dedicated cluster. Every path starts from TEE-ready infrastructure instead of a raw GPU box.
Trial ready

Proven confidential inference and fine-tuning capacity.
Memory
80GB HBM3
Bandwidth
3.35 TB/s
Region
US-West
Scale
1-2 GPUs
On-demand
$3.08/GPU/hr
24h minimum
Slot
$2.38/GPU/hr
reserved
Slot ready

High-memory runtime for larger private model jobs.
Memory
141GB HBM3e
Bandwidth
4.8 TB/s
Region
US-West / India
Scale
1-8 GPUs
On-demand
$4.80/GPU/hr
24h minimum
Slot
$3.20/GPU/hr
reserved
Quote now

Blackwell Ultra confidential capacity for frontier inference.
Memory
288GB HBM3e
Bandwidth
8 TB/s
Region
US-East / US-West
Scale
1-8 GPUs
1-month
$6.50/GPU/hr
30d minimum
Slot
$5.60/GPU/hr
reserved
Prices include Intel TDX + NVIDIA confidential computing readiness. Volume and enterprise pricing are quoted by workload.
relative index
1x
1.9x
3.2x
LLM 推理
model + KV cache
80GB
141GB
288GB
GPU 内存
feed batches
3.35TB/s
4.8TB/s
8TB/s
内存带宽

NVIDIA H100
80GB HBM3

NVIDIA H200
141GB HBM3e

NVIDIA B300
288GB HBM3e
GPU 对比
在报价前先比较容量形态。H100 适合快速试用,H200 提供更大显存余量,B300 则是面向前沿推理和专属集群的 Blackwell Ultra 路径。
实际吞吐量取决于模型、批大小、精度和运行时。Phala 会将 GPU 与隐私虚拟机路径、GPU CC 就绪度和证明操作一起报价。
GPU 云示意图
市场视图应当让购买路径一目了然:试用、预留,然后扩展到带有 TEE 就绪状态的专属集群。

H100
80GB HBM3from
$3.08/hr

H200
141GB HBM3efrom
$4.80/hr

B300
288GB HBM3efrom
$6.50/hr
已验证的
CVM 运行时
已验证的
GPU CC 模式
已验证的
双重证明
GPU TEE 证明路径
只有当整个路径——运行时、GPU 模式和证据采集——端到端可验证时,GPU 隔离才有意义。Phala 将三者一起交付。
01
Docker 工作负载在带 GPU 直通的 Intel TDX 机密虚拟机中运行。运行时在工作负载启动前会被封存以对抗操作员,并由固件度量。
02
NVIDIA Confidential Computing 将模型权重、激活值和 KV cache 封装在受保护的 GPU 内存中。GPU 与 CPU TEE 一起执行计算隔离。
03
Intel TDX 和 NVIDIA 都会输出签名 quote。Phala 将两者收集并通过一个验证器对外呈现,让 CVM 和 GPU 一起证明自身。
AI 解决方案路径
隐私模型端点是第一个入口点。同样的隐私原语也适用于代理、数据工作流和训练。
提供 OpenAI 兼容的模型调用,提示词、输出和客户上下文都需要在使用中加密保护。
128K
$0.27/M input
256K
$0.40/M input
128K
$0.15/M input
128K
$0.10/M input
200K
$3.00/M input
1M
$1.25/M input
在可验证的运行时中运行代理的密钥、工具、记忆和操作,而不是放在可见的自动化云中。
在保持数据集、梯度、检查点和评估轨迹处于边界内的同时,基于专有数据调整模型。
private training run
01
sealed
02
running
03
private
04
verified
loss curve
proof attached
attestation.json
将模型移动到敏感记录旁,在不向模型运营方暴露原始数据的情况下返回已批准的输出。
source
EHR data
source
Customer records
source
Internal docs
TEE clean room
approved output