Flop
per.io
GPU Specification Sheet
August 11, 2026
flopper.io/gpu/huawei-atlas-350
Huawei Atlas 350
Da Vinci v3
Custom
2026
Overview
—
FP32 TFLOPS
112 GB
VRAM HiBL 1.0
1.4 TB/s
Memory Bandwidth
600 W
TDP
Performance Metrics
Precision
Peak TFLOPS
Bits
Type
FP6
804.0
—
Standard
INT8
804.0
8
Standard
FP8
804.0
8
Standard
FP4
1561.0
4
Standard
FP16
425.0
16
Standard
BF16
425.0
16
Standard
Memory Specifications
Capacity
112 GB
Type
HiBL 1.0
Bandwidth
1.4 TB/s
Interface Width
—
Power & Efficiency
TDP
600 W
Est. Max Power
690 W
FP32 Efficiency
—
FP16 Efficiency
0.71 TFLOPS/W
Hardware Details
Vendor:
Huawei
Architecture:
Da Vinci v3
Process Node:
—
Form Factor:
Custom
Launch Year:
2026
Interconnect
GPU-to-GPU:
Unified Bus 2.0