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: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 256, 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}
0.070753
536,870,912
8,650,752
62.060606
4
{"M": 256, "N": 256, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 256, 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}
0.07528
536,870,912
8,650,752
62.060606
4
{"M": 256, "N": 256, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 256, 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}
0.040468
536,870,912
8,650,752
62.060606
4
{"M": 256, "N": 256, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 256, 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}
0.07463
536,870,912
8,650,752
62.060606
4
{"M": 256, "N": 256, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.066928
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.059427
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.046331
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.048806
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.05094
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.065531
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.044816
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.053212
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.036184
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.04187
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.025658
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 4096) x (4096, 256) -> (256, 256) C = torch.matmul(A, B) return C A = torch.randn(256, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 256, 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}
0.069598
536,870,912
4,325,376
124.121212
2
{"M": 256, "N": 256, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.054041
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.056804
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.027805
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.055426
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.049025
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.02176
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.061708
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, 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}
0.056257
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.048952
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.062919
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.039408
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float32, device='cuda') B = torch.randn(64, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.055587
16,777,216
720,896
23.272727
4
{"M": 256, "N": 512, "K": 64, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA T4
{"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4}
0.054513
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA V100
{"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6}
0.048369
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A10G
{"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6}
0.068172
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 40GB
{"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.037761
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 80GB
{"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.020179
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L4
{"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48}
0.044479
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L40S
{"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96}
0.022417
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 3090
{"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6}
0.073924
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.05559
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.041063
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.037128
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 64) x (64, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 64, dtype=torch.float16, device='cuda') B = torch.randn(64, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.040244
16,777,216
360,448
46.545455
2
{"M": 256, "N": 512, "K": 64, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.066175
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.057067
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.039799
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.057836
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.049007
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.046644
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.055041
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, 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}
0.04911
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.054058
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.033636
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.02726
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float32, device='cuda') B = torch.randn(256, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.031091
67,108,864
1,310,720
51.2
4
{"M": 256, "N": 512, "K": 256, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA T4
{"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4}
0.029728
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA V100
{"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6}
0.041497
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A10G
{"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6}
0.041968
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 40GB
{"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.05549
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 80GB
{"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.057391
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L4
{"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48}
0.033491
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L40S
{"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96}
0.029431
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 3090
{"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6}
0.056657
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.031949
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.057347
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.031185
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 256) x (256, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 256, dtype=torch.float16, device='cuda') B = torch.randn(256, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.03984
67,108,864
655,360
102.4
2
{"M": 256, "N": 512, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.072557
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.065018
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.064538
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.060519
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.064574
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.056928
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA 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}
0.065113
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 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}
0.035065
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.052003
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.041994
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.043432
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 512, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.039397
134,217,728
2,097,152
64
4
{"M": 256, "N": 512, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA T4
{"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4}
0.043117
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA V100
{"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6}
0.043935
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A10G
{"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6}
0.058069
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 40GB
{"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.025884
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 80GB
{"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.032513
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L4
{"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48}
0.06051
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L40S
{"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96}
0.062805
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 3090
{"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6}
0.056255
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.030379
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.051599
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.030624
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 512) x (512, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 512, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.044446
134,217,728
1,048,576
128
2
{"M": 256, "N": 512, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.099794
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.046316
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.066331
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.080128
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.089713
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.052557
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.050983
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.070308
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.028869
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.062725
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.032504
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (256, 1024) x (1024, 512) -> (256, 512) C = torch.matmul(A, B) return C A = torch.randn(256, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 512, 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}
0.064714
268,435,456
3,670,016
73.142857
4
{"M": 256, "N": 512, "K": 1024, "dtype": "float32"}