DEMO MODESimulated Compliance Environment • No Live Production Data
CEO Hub
HYBA

MacQuantumCuda

Cross-Vendor GPU Acceleration · Apple Silicon + NVIDIA CUDA · Substrate-Agnostic Quantum

107.0 TFLOPS FleetMetal / MLX / CUDA
Fleet TFLOPS

107.0 FP32

UMA Bandwidth

3K GB/s

Avg Utilization

61%

Avg Temp

53°C

Active GPUs

5 units

Total Memory

640 GB UMA

Max Power

1040 W

NVIDIA Backends

5 benchmarked

active
Apple M2 Ultra
Metal/MLX · 76 cores · 295W TDP
Memory192 GB
Bandwidth819 GB/s
Utilization67%
TFLOPS: 27.258°C
active
Apple M2 Ultra
Metal/MLX · 76 cores · 295W TDP
Memory192 GB
Bandwidth819 GB/s
Utilization72%
TFLOPS: 27.261°C
active
Apple M3 Max
Metal/MLX · 40 cores · 195W TDP
Memory128 GB
Bandwidth546 GB/s
Utilization45%
TFLOPS: 18.452°C
active
Apple M3 Max
Metal/MLX · 40 cores · 195W TDP
Memory128 GB
Bandwidth546 GB/s
Utilization38%
TFLOPS: 18.449°C
active
Apple Neural Engine
ANE · 32 cores · 60W TDP
MemoryShared
Bandwidth200 GB/s
Utilization82%
TFLOPS: 15.844°C