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, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 512, 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} | 0.067178 | 536,870,912 | 5,242,880 | 102.4 | 4 | {"M": 1024, "N": 512, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 512, 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} | 0.059136 | 536,870,912 | 5,242,880 | 102.4 | 4 | {"M": 1024, "N": 512, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 512, 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} | 0.05996 | 536,870,912 | 5,242,880 | 102.4 | 4 | {"M": 1024, "N": 512, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 512, 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} | 0.065522 | 536,870,912 | 5,242,880 | 102.4 | 4 | {"M": 1024, "N": 512, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.055782 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.069998 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.058065 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.038083 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.04561 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.057221 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.031426 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.049241 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.054014 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.026433 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.061024 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 512) x (512, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 512, 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} | 0.066629 | 536,870,912 | 2,621,440 | 204.8 | 2 | {"M": 1024, "N": 512, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.232654 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.151643 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.122428 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.11722 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.101327 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.105257 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.054607 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.060415 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.048765 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.057796 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.067416 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.071748 | 1,073,741,824 | 8,388,608 | 128 | 4 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.048825 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.053175 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.065419 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.053743 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.052891 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.066635 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.053586 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.083437 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.057291 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.037141 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.042495 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 1024) x (1024, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.050901 | 1,073,741,824 | 4,194,304 | 256 | 2 | {"M": 1024, "N": 512, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.420813 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.208102 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.140768 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.186728 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.164996 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.119455 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.068457 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.09109 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.083565 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.075315 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.092819 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.116511 | 2,147,483,648 | 14,680,064 | 146.285714 | 4 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.099529 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.068759 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.0615 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.055264 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.067849 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.061844 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.056907 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.07969 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.056564 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.043888 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.059926 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 2048) x (2048, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.071951 | 2,147,483,648 | 7,340,032 | 292.571429 | 2 | {"M": 1024, "N": 512, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.945825 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.386959 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.242059 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.322576 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.324516 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.198631 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.087738 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.196645 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.091454 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.107925 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.164019 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.166991 | 4,294,967,296 | 27,262,976 | 157.538462 | 4 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.139911 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.10021 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.138696 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.048276 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.07412 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.11613 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.090348 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.12401 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.058565 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.031875 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.032893 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 4096) x (4096, 512) -> (1024, 512)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 512, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchro... | 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} | 0.129587 | 4,294,967,296 | 13,631,488 | 315.076923 | 2 | {"M": 1024, "N": 512, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.083466 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.058517 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.055113 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.060444 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.063227 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.073416 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.079596 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.055554 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.03095 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.058485 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.059254 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (1024, 64) x (64, 1024) -> (1024, 1024)
C = torch.matmul(A, B)
return C
A = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 1024, 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} | 0.028394 | 134,217,728 | 4,718,592 | 28.444444 | 4 | {"M": 1024, "N": 1024, "K": 64, "dtype": "float32"} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.