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: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.091972 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.081206 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.08121 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.069961 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.058136 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.064413 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.07812 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.092363 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.078242 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.068526 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.043446 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.022832 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.038129 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.048566 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.05992 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.037588 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.05043 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.06075 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.038639 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 512) x (512, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 512, dtype=torch.float16, device='cuda')
B = torch.randn(512, 64, 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} | 0.045074 | 536,870,912 | 9,502,720 | 56.496552 | 2 | {"M": 8192, "N": 64, "K": 512, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.252061 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.138666 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.158262 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.13479 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.138628 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.131783 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.100172 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.116248 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.091512 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.072968 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.097492 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float32, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.111379 | 1,073,741,824 | 35,913,728 | 29.89781 | 4 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.121026 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.051617 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.088889 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.068172 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.048503 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.113503 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.056115 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.069933 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.03088 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.022671 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.066476 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 1024) x (1024, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 1024, dtype=torch.float16, device='cuda')
B = torch.randn(1024, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.095238 | 1,073,741,824 | 17,956,864 | 59.79562 | 2 | {"M": 8192, "N": 64, "K": 1024, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.427645 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.238334 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.250188 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.165877 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.210924 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.359707 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.122765 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.178943 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.143753 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.090876 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.111019 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.194513 | 2,147,483,648 | 69,730,304 | 30.796992 | 4 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.227231 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.093972 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.107387 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.079954 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.095388 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.092378 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.076464 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.092529 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.100498 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.051304 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.066014 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 2048) x (2048, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.085848 | 2,147,483,648 | 34,865,152 | 61.593985 | 2 | {"M": 8192, "N": 64, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.822416 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.413019 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.388941 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.404517 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.328919 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.62278 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.254663 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.219562 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.335108 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.1514 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.159146 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.311046 | 4,294,967,296 | 137,363,456 | 31.267176 | 4 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.38478 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.141124 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.225542 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.132451 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.100693 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.462338 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.115084 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.098293 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.148673 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.062008 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.097785 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 4096) x (4096, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 64, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.synchroniz... | 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} | 0.174653 | 4,294,967,296 | 68,681,728 | 62.534351 | 2 | {"M": 8192, "N": 64, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.099219 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.071831 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.059037 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.060504 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.042784 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.050995 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.065132 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 256) -> (8192, 256)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 256, 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} | 0.052141 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 256, "K": 64, "dtype": "float32"} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.