Product overview
Purpose-built accelerator for enterprise AI workloads
Designed for dense inference, model adaptation, and private AI infrastructure where predictable availability, localized supply, and software compatibility are critical.
Architecture
TPU Архитектура
Memory & Performance
High-bandwidth memory profile for AI inference and training.
VRAM capacity96
Memory typeHBM3e
Memory bandwidth1200
Interconnect typeYHLink
Interconnect speed1200
Architecture
Compute architecture and software execution model.
ArchitectureTPU Архитектура
Compute units-
Power & Thermal
Data center integration requirements.
Thermal design power-
CoolingПассивное
Form factorMezzanine Module
Pixel Rate-
Texture Rate-
Benchmarks
Peak theoretical compute for common AI precisions.
FP64
INT4 TOPS 2048
FP32
128
FP16
-
TF32
-
BF16 Tensor
512
FP8 Tensor
1024
INT8 Tensor
1024
Compatibility
Interfaces, frameworks, and deployment environment.
PCIe interfacePCIe 5.0x16
Video encoding-
Video decoding-
Physical Dimensions
Card dimensions for server platforms.
Slots-
Length- mm
Height- mm
Width- mm
Pricing
On request
Volume pricing available for cluster deployments and pilot batches.
Documentation