GPUs - HMC

GPUs

Intel

Arc A570M

Shaders

2,048

Boost

1.3GHz

Memory

8GB

TDP

75W

FP32

5.33 TFLOPS—

Peak

170.4 TOPSINT4 Tensor—

—

Intel

Arc A530M

Shaders

1,536

Boost

1.2GHz

Memory

4GB

TDP

65W

FP32

3.69 TFLOPS—

Peak

118 TOPSINT4 Tensor—

—

AMD

Radeon RX 7900 GRE

Shaders

5,120

Boost

2.25GHz

Memory

16GB

TDP

260W

FP32

45.98 TFLOPS—

Peak

183.9 TOPSINT4 Tensor—

$550

Intel

Arc Pro A60

Shaders

2,048

Boost

2.05GHz

Memory

12GB

TDP

130W

FP32

8.4 TFLOPS—

Peak

268.7 TOPSINT4 Tensor—

—

Intel

Arc Pro A60M

Shaders

2,048

Boost

1.3GHz

Memory

8GB

TDP

95W

FP32

5.33 TFLOPS—

Peak

170.4 TOPSINT4 Tensor—

—

Apple

76-Core M2 Ultra

Shaders

9,728

Boost

1.4GHz

Memory

Shared

TDP

Shared

FP32

27.2 TFLOPS—

Peak

27.2 TFLOPSFP16—

—

Apple

60-Core M2 Ultra

Shaders

7,680

Boost

1.4GHz

Memory

Shared

TDP

Shared

FP32

21.47 TFLOPS—

Peak

21.47 TFLOPSFP16—

—

Qualcomm

Adreno 613

Shaders

128

Boost

955MHz

Memory

Shared

TDP

FP32

244.5 GFLOPS—

Peak

489 GFLOPSFP16—

—

AMD

Radeon RX 7600

Shaders

2,048

Boost

2.25GHz

Memory

8GB

TDP

165W

FP32

18.43 TFLOPS—

Peak

73.73 TOPSINT4 Tensor—

$270

NVIDIA

GeForce RTX 4060 Ti

Shaders

4,352

Boost

2.54GHz

Memory

10GB

TDP

160W

FP32

22.11 TFLOPS—

Peak

707.5 TOPSINT4 Tensor Sparse—

$400

NVIDIA

GeForce RTX 4060

Shaders

3,072

Boost

2.46GHz

Memory

8GB

TDP

115W

FP32

15.11 TFLOPS—

Peak

483.7 TOPSINT4 Tensor Sparse—

$300

AMD

Radeon 780M 2700MHz

Shaders

768

Boost

2.7GHz

Memory

Shared

TDP

Shared

FP32

4.15 TFLOPS—

Peak

8.29 TFLOPSFP16—

—

AMD

Radeon 740M 2800MHz

Shaders

256

Boost

2.8GHz

Memory

Shared

TDP

Shared

FP32

1.43 TFLOPS—

Peak

2.87 TFLOPSFP16—

—

AMD

Radeon 740M 2700MHz

Shaders

256

Boost

2.7GHz

Memory

Shared

TDP

Shared

FP32

1.38 TFLOPS—

Peak

2.77 TFLOPSFP16—

—

AMD

Radeon 760M 2600MHz

Shaders

384

Boost

2.6GHz

Memory

Shared

TDP

Shared

FP32

2 TFLOPS—

Peak

3.99 TFLOPSFP16—

—

AMD

Radeon 740M 2500MHz

Shaders

256

Boost

2.5GHz

Memory

Shared

TDP

Shared

FP32

1.28 TFLOPS—

Peak

2.56 TFLOPSFP16—

—

NVIDIA

RTX A1000

Shaders

2,307

Boost

1.46GHz

Memory

8GB

TDP

50W

FP32

6.75 TFLOPS—

Peak

215.6 TOPSINT4 Tensor Sparse—

—

NVIDIA

RTX A400

Shaders

768

Boost

1.76GHz

Memory

4GB

TDP

50W

FP32

2.7 TFLOPS—

Peak

86.51 TOPSINT4 Tensor Sparse—

—

AMD

Radeon PRO W7900

Shaders

6,144

Boost

2.5GHz

Memory

48GB

TDP

295W

FP32

61.32 TFLOPS—

Peak

245.3 TOPSINT4 Tensor—

$4,000

AMD

Radeon PRO W7800

Shaders

4,480

Boost

2.52GHz

Memory

32GB

TDP

260W

FP32

45.25 TFLOPS—

Peak

181 TOPSINT4 Tensor—

$2,500

NVIDIA

GeForce RTX 4070

Shaders

5,888

Boost

2.48GHz

Memory

12GB

TDP

200W

FP32

29.15 TFLOPS—

Peak

932.7 TOPSINT4 Tensor Sparse—

$600

Intel

Max 1100

Shaders

7,168

Boost

1.55GHz

Memory

48GB

TDP

300W

FP32

22.22 TFLOPS—

Peak

711.1 TOPSINT8 Tensor—

—

NVIDIA

L4

Shaders

7,680

Boost

2.04GHz

Memory

24GB

TDP

72W

FP32

31.33 TFLOPS—

Peak

1 POPSINT4 Tensor Sparse—

—

NVIDIA

RTX 5000 Max-Q

Shaders

9,728

Boost

1.68GHz

Memory

16GB

TDP

120W

FP32

32.69 TFLOPS—

Peak

1.05 POPSINT4 Tensor Sparse—

—

NVIDIA

RTX 5000 Laptop

Shaders

9,728

Boost

2.19GHz

Memory

16GB

TDP

150W

FP32

42.61 TFLOPS—

Peak

1.36 POPSINT4 Tensor Sparse—

—

Page 11