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: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 16.760518 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 7.103613 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 3.327772 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 4.116134 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 5.121107 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 2.941435 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 1.036258 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 1.903509 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 1.012121 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 1.638871 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 1.908023 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float32, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 2.112478 | 68,719,476,736 | 134,217,728 | 512 | 4 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 1.157372 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 0.972032 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 1.467707 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.384255 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.283136 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 0.879719 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.507976 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 1.194055 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 0.452356 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.110183 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.188673 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 2048) x (2048, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 2048, dtype=torch.float16, device='cuda')
B = torch.randn(2048, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 1.00294 | 68,719,476,736 | 67,108,864 | 1,024 | 2 | {"M": 4096, "N": 4096, "K": 2048, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 28.481373 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 11.838301 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 6.304511 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 8.837141 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 10.520565 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 6.127646 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 2.084911 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 4.164314 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 1.978084 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 2.689402 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 4.775562 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float32, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 5.017624 | 137,438,953,472 | 201,326,592 | 682.666667 | 4 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA T4 | {"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4} | 3.339065 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA V100 | {"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6} | 1.5624 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A10G | {"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6} | 2.906222 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 40GB | {"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.716499 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA A100 80GB | {"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40} | 0.719448 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L4 | {"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48} | 2.051938 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA L40S | {"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96} | 0.867441 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 3090 | {"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6} | 2.834046 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX 4090 | {"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72} | 1.208293 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 SXM | {"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.184808 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA H100 PCIe | {"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50} | 0.259452 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (4096, 4096) x (4096, 4096) -> (4096, 4096)
C = torch.matmul(A, B)
return C
A = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
B = torch.randn(4096, 4096, dtype=torch.float16, device='cuda')
C = matmul_kernel(A, B)
torch.cuda.sync... | matmul | NVIDIA RTX A6000 | {"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6} | 2.529028 | 137,438,953,472 | 100,663,296 | 1,365.333333 | 2 | {"M": 4096, "N": 4096, "K": 4096, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 64, 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} | 0.044768 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 64, 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} | 0.050176 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 64, 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} | 0.068316 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 64, 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} | 0.040862 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.037086 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.048925 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.057011 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.032141 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.068643 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.052037 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.018056 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float32, device='cuda')
B = torch.randn(64, 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} | 0.038478 | 67,108,864 | 4,210,688 | 15.937743 | 4 | {"M": 8192, "N": 64, "K": 64, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.066416 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.051839 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.043946 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.027188 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.022177 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.065567 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.058006 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.023361 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.045877 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.032384 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.053077 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 64) x (64, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 64, dtype=torch.float16, device='cuda')
B = torch.randn(64, 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} | 0.048837 | 67,108,864 | 2,105,344 | 31.875486 | 2 | {"M": 8192, "N": 64, "K": 64, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 64, 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} | 0.068475 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 64, 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} | 0.059442 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 64, 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} | 0.064848 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 64, 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} | 0.056232 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.070449 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.054553 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.046079 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.061554 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.05894 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.034539 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.056789 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float32, device='cuda')
B = torch.randn(256, 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} | 0.052251 | 268,435,456 | 10,551,296 | 25.440994 | 4 | {"M": 8192, "N": 64, "K": 256, "dtype": "float32"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.068512 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.059103 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.045399 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.034908 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.05635 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.086201 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.037311 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.068648 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.058725 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.049133 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.058465 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "dtype": "float16"} |
import torch
def matmul_kernel(A, B):
# Matrix multiplication: (8192, 256) x (256, 64) -> (8192, 64)
C = torch.matmul(A, B)
return C
A = torch.randn(8192, 256, dtype=torch.float16, device='cuda')
B = torch.randn(256, 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} | 0.060467 | 268,435,456 | 5,275,648 | 50.881988 | 2 | {"M": 8192, "N": 64, "K": 256, "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.float32, device='cuda')
B = torch.randn(512, 64, 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} | 0.143475 | 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 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} | 0.087703 | 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 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} | 0.090699 | 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 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} | 0.074698 | 536,870,912 | 19,005,440 | 28.248276 | 4 | {"M": 8192, "N": 64, "K": 512, "dtype": "float32"} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.