code large_stringlengths 193 899 | workload_type large_stringclasses 15
values | gpu_name large_stringclasses 12
values | gpu_features large_stringclasses 12
values | runtime_ms float64 0.01 4.35k | flops float64 30 21,045B | memory_bytes int64 160 13.2B | arithmetic_intensity float64 0 6.37k | dtype_bytes int64 2 4 | workload_params large_stringlengths 11 79 |
|---|---|---|---|---|---|---|---|---|---|
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchron... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.05902 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 2048, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchron... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.02267 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 2048, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchron... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.065453 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 2048, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchron... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.093581 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 2048, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.899234 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.432867 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.237724 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.343271 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.340493 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.241823 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.089844 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.232104 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.102569 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.138971 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.129151 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.22305 | 4,294,967,296 | 20,971,520 | 204.8 | 4 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.168088 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.086124 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.130163 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.0624 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.05368 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.122528 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.053918 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.153878 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.071769 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.050429 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.047087 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.111484 | 4,294,967,296 | 10,485,760 | 409.6 | 2 | {"M": 1024, "N": 2048, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 1.305315 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.848777 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.342461 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.507181 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.651055 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.359241 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.164414 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.434556 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.193012 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.191557 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.36918 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.335618 | 8,589,934,592 | 33,554,432 | 256 | 4 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.178722 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.161153 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.173325 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.100217 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.111672 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.104703 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.112251 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.177909 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.118305 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.036812 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.060027 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.146639 | 8,589,934,592 | 16,777,216 | 512 | 2 | {"M": 1024, "N": 2048, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 2.999737 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 1.514186 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.670228 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 1.426368 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 1.030567 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.994823 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.333571 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.609199 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.329761 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.28987 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.410108 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.779265 | 17,179,869,184 | 58,720,256 | 292.571429 | 4 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.381979 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.236559 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.445164 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.116753 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.131344 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.193588 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.223782 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.442197 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.191366 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.060675 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.060289 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 2048) -> (1024, 2048)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.397363 | 17,179,869,184 | 29,360,128 | 585.142857 | 2 | {"M": 1024, "N": 2048, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.152153 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.100812 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.086505 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.097391 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.066556 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.12765 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.068191 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.058883 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.041797 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.03722 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.053692 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.115165 | 536,870,912 | 18,087,936 | 29.681159 | 4 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 0.081256 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.068606 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.04789 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.030519 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.039178 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.055527 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.044836 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.056399 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.069244 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.019442 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.061229 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 4096) -> (1024, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchronize(... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 0.053951 | 536,870,912 | 9,043,968 | 59.362319 | 2 | {"M": 1024, "N": 4096, "K": 64, "dtype": "float16"} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.