GPUs - HMC

NVIDIA

RTX PRO 3000 Blackwell Laptop

Shaders

5,888

Boost

2.47GHz

Memory

12GB

TDP

140W

FP32

29.1 TFLOPS—

Peak

992 TFLOPSNVFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 4000 Blackwell Laptop

Shaders

7,680

Boost

2.52GHz

Memory

16GB

TDP

175W

FP32

38.7 TFLOPS—

Peak

1.33 PFLOPSNVFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 5000 Blackwell Laptop

Shaders

10,496

Boost

2.37GHz

Memory

24GB

TDP

175W

FP32

49.8 TFLOPS—

Peak

1.82 PFLOPSNVFP4 Tensor Sparse—

—

NVIDIA

GB300

Shaders

20,480

Boost

2.04GHz

Memory

288GB

TDP

2800W

FP32

166.7 TFLOPS(2x 83.33 TFLOPS)—

Peak

40 PFLOPSMXFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 6000 Blackwell

Shaders

24,064

Boost

2.62GHz

Memory

96GB

TDP

600W

FP32

126 TFLOPS—

Peak

4.03 PFLOPSFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 6000 Blackwell Max-Q

Shaders

24,064

Boost

2.28GHz

Memory

96GB

TDP

300W

FP32

109.7 TFLOPS—

Peak

3.51 PFLOPSFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 6000 Blackwell Server

Shaders

24,064

Boost

2.43GHz

Memory

96GB

TDP

600W

FP32

117 TFLOPS—

Peak

3.74 PFLOPSFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 5000 Blackwell

Shaders

14,080

Boost

2.62GHz

Memory

48GB

TDP

300W

FP32

73.69 TFLOPS—

Peak

2.36 PFLOPSFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 4500 Blackwell

Shaders

10,496

Boost

2.62GHz

Memory

32GB

TDP

200W

FP32

54.94 TFLOPS—

Peak

1.76 PFLOPSFP4 Tensor Sparse—

—

NVIDIA

RTX PRO 4000 Blackwell

Shaders

8,960

Boost

2.45GHz

Memory

24GB

TDP

140W

FP32

43.94 TFLOPS—

Peak

1.41 PFLOPSFP4 Tensor Sparse—

—

ARM

Mali-G57 MP2 880MHz

Shaders

64

Boost

880MHz

Memory

Shared

TDP

Shared

FP32

112.6 GFLOPS—

Peak

225.3 GFLOPSFP16—

—

ARM

Mali-G57 MP2 1100MHz

Shaders

64

Boost

1.1GHz

Memory

Shared

TDP

Shared

FP32

140.8 GFLOPS—

Peak

281.6 GFLOPSFP16—

—

Apple

8-Core M4-FL

Shaders

1,024

Boost

1.47GHz

Memory

Shared

TDP

Shared

FP32

3.01 TFLOPS—

Peak

3.01 TFLOPSFP16—

—

Apple

80-Core M3 Ultra

Shaders

10,240

Boost

1.38GHz

Memory

Shared

TDP

Shared

FP32

28.26 TFLOPS—

Peak

28.26 TFLOPSFP16—

—

Apple

60-Core M3 Ultra

Shaders

7,680

Boost

1.38GHz

Memory

Shared

TDP

Shared

FP32

21.2 TFLOPS—

Peak

21.2 TFLOPSFP16—

—

Apple

4-Core A16

Shaders

512

Boost

1.34GHz

Memory

Shared

TDP

Shared

FP32

1.37 TFLOPS—

Peak

1.37 TFLOPSFP16—

—

Apple

9-Core M3-FL

Shaders

1,152

Boost

1.34GHz

Memory

Shared

TDP

Shared

FP32

3.08 TFLOPS—

Peak

3.08 TFLOPSFP16—

—

AMD

Radeon RX 7650 GRE

Shaders

2,048

Boost

2.69GHz

Memory

8GB

TDP

165W

FP32

22.08 TFLOPS—

Peak

88.31 TOPSINT4 Tensor—

$280

ARM

Mali-G615 MP2 1047MHz

Shaders

256

Boost

1.05GHz

Memory

Shared

TDP

Shared

FP32

536.1 GFLOPS—

Peak

1.07 TFLOPSFP16—

—

ARM

Mali-G615 MP2 1047MHz

Shaders

256

Boost

1.05GHz

Memory

Shared

TDP

Shared

FP32

536.1 GFLOPS—

Peak

1.07 TFLOPSFP16—

—

ARM

Mali-G57 MP2 1070MHz

Shaders

64

Boost

1.07GHz

Memory

Shared

TDP

Shared

FP32

137 GFLOPS—

Peak

273.9 GFLOPSFP16—

—

Apple

4-Core A18

Shaders

512

Boost

1.34GHz

Memory

Shared

TDP

Shared

FP32

1.37 TFLOPS—

Peak

1.37 TFLOPSFP16—

—

NVIDIA

Tegra Switch 2

Shaders

1,536

Boost

1.01GHz

Memory

Shared

TDP

Shared

FP32

3.09 TFLOPS—

Peak

49.5 TOPSINT4 Tensor—

—

NVIDIA

GeForce RTX 5090 D v2

Shaders

21,760

Boost

2.41GHz

Memory

24GB

TDP

575W

FP32

104.8 TFLOPS—

Peak

2.38 PFLOPSFP4 Tensor Sparse—

$2,300

NVIDIA

GeForce RTX 5090 D

Shaders

21,760

Boost

2.41GHz

Memory

32GB

TDP

575W

FP32

104.8 TFLOPS—

Peak

2.38 PFLOPSFP4 Tensor Sparse—

$2,300

Page 6