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
DaVinci
Memory & Performance
High-bandwidth memory profile for AI inference and training.
VRAM capacity48
Memory typeLPDDR4x
Memory bandwidth205
Interconnect type-
Interconnect speed64
Architecture
Compute architecture and software execution model.
Architecture DaVinci
Compute units-
Power & Thermal
Data center integration requirements.
Thermal design power72
CoolingПассивное
Form factorPCIe
Pixel Rate-
Texture Rate-
Benchmarks
Peak theoretical compute for common AI precisions.
FP64
-
FP32
-
FP16
70
TF32
NA
BF16 Tensor
70
FP8 Tensor
NA
INT8 Tensor
140
Compatibility
Interfaces, frameworks, and deployment environment.
PCIe interfacePCIe 4.0x16
Video encoding24x1080p30fps
Video decoding128x1080p30fps
Physical Dimensions
Card dimensions for server platforms.
Slots1
Length- mm
Height- mm
Width- mm
Pricing
On request
Volume pricing available for cluster deployments and pilot batches.
Documentation