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: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.039046
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.042592
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.051827
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.071077
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.05491
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.044699
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.056595
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.032767
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.053522
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.045562
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.03629
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 256) x (256, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.021965
4,194,304
114,688
36.571429
2
{"M": 128, "N": 64, "K": 256, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.039293
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.032293
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.033322
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 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}
0.043094
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 80GB
{"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.027709
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L4
{"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48}
0.053833
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L40S
{"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96}
0.028756
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 3090
{"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6}
0.023705
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.061783
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.050804
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.048593
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float32, device='cuda') B = torch.randn(512, 64, dtype=torch.float32, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.048903
8,388,608
425,984
19.692308
4
{"M": 128, "N": 64, "K": 512, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA T4
{"gpu_name": "NVIDIA T4", "cuda_cores": 2560, "tensor_cores": 320, "memory_gb": 16, "memory_bandwidth_gbps": 320, "base_clock_mhz": 585, "boost_clock_mhz": 1590, "sm_count": 40, "fp32_tflops": 8.1, "fp16_tflops": 65, "tdp_watts": 70, "compute_capability": 7.5, "l2_cache_mb": 4}
0.058018
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA V100
{"gpu_name": "NVIDIA V100", "cuda_cores": 5120, "tensor_cores": 640, "memory_gb": 32, "memory_bandwidth_gbps": 900, "base_clock_mhz": 1230, "boost_clock_mhz": 1530, "sm_count": 80, "fp32_tflops": 15.7, "fp16_tflops": 125, "tdp_watts": 300, "compute_capability": 7.0, "l2_cache_mb": 6}
0.026385
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A10G
{"gpu_name": "NVIDIA A10G", "cuda_cores": 9216, "tensor_cores": 288, "memory_gb": 24, "memory_bandwidth_gbps": 600, "base_clock_mhz": 885, "boost_clock_mhz": 1695, "sm_count": 80, "fp32_tflops": 31.2, "fp16_tflops": 62.5, "tdp_watts": 150, "compute_capability": 8.6, "l2_cache_mb": 6}
0.024559
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 40GB
{"gpu_name": "NVIDIA A100 40GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 40, "memory_bandwidth_gbps": 1555, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.045096
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA A100 80GB
{"gpu_name": "NVIDIA A100 80GB", "cuda_cores": 6912, "tensor_cores": 432, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 765, "boost_clock_mhz": 1410, "sm_count": 108, "fp32_tflops": 19.5, "fp16_tflops": 312, "tdp_watts": 400, "compute_capability": 8.0, "l2_cache_mb": 40}
0.028193
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L4
{"gpu_name": "NVIDIA L4", "cuda_cores": 7424, "tensor_cores": 232, "memory_gb": 24, "memory_bandwidth_gbps": 300, "base_clock_mhz": 795, "boost_clock_mhz": 2040, "sm_count": 58, "fp32_tflops": 30.3, "fp16_tflops": 121, "tdp_watts": 72, "compute_capability": 8.9, "l2_cache_mb": 48}
0.045554
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA L40S
{"gpu_name": "NVIDIA L40S", "cuda_cores": 18176, "tensor_cores": 568, "memory_gb": 48, "memory_bandwidth_gbps": 864, "base_clock_mhz": 1110, "boost_clock_mhz": 2520, "sm_count": 142, "fp32_tflops": 91.6, "fp16_tflops": 183.2, "tdp_watts": 350, "compute_capability": 8.9, "l2_cache_mb": 96}
0.04343
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 3090
{"gpu_name": "NVIDIA RTX 3090", "cuda_cores": 10496, "tensor_cores": 328, "memory_gb": 24, "memory_bandwidth_gbps": 936, "base_clock_mhz": 1395, "boost_clock_mhz": 1695, "sm_count": 82, "fp32_tflops": 35.6, "fp16_tflops": 71, "tdp_watts": 350, "compute_capability": 8.6, "l2_cache_mb": 6}
0.058274
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX 4090
{"gpu_name": "NVIDIA RTX 4090", "cuda_cores": 16384, "tensor_cores": 512, "memory_gb": 24, "memory_bandwidth_gbps": 1008, "base_clock_mhz": 2235, "boost_clock_mhz": 2520, "sm_count": 128, "fp32_tflops": 82.6, "fp16_tflops": 165.2, "tdp_watts": 450, "compute_capability": 8.9, "l2_cache_mb": 72}
0.039695
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 SXM
{"gpu_name": "NVIDIA H100 SXM", "cuda_cores": 16896, "tensor_cores": 528, "memory_gb": 80, "memory_bandwidth_gbps": 3350, "base_clock_mhz": 1095, "boost_clock_mhz": 1830, "sm_count": 132, "fp32_tflops": 67, "fp16_tflops": 989, "tdp_watts": 700, "compute_capability": 9.0, "l2_cache_mb": 50}
0.038424
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA H100 PCIe
{"gpu_name": "NVIDIA H100 PCIe", "cuda_cores": 14592, "tensor_cores": 456, "memory_gb": 80, "memory_bandwidth_gbps": 2039, "base_clock_mhz": 1095, "boost_clock_mhz": 1620, "sm_count": 114, "fp32_tflops": 48, "fp16_tflops": 756, "tdp_watts": 350, "compute_capability": 9.0, "l2_cache_mb": 50}
0.048281
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 512) x (512, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 512, dtype=torch.float16, device='cuda') B = torch.randn(512, 64, dtype=torch.float16, device='cuda') C = matmul_kernel(A, B) torch.cuda.synchronize()
matmul
NVIDIA RTX A6000
{"gpu_name": "NVIDIA RTX A6000", "cuda_cores": 10752, "tensor_cores": 336, "memory_gb": 48, "memory_bandwidth_gbps": 768, "base_clock_mhz": 1410, "boost_clock_mhz": 1860, "sm_count": 84, "fp32_tflops": 38.7, "fp16_tflops": 77.4, "tdp_watts": 300, "compute_capability": 8.6, "l2_cache_mb": 6}
0.020214
8,388,608
212,992
39.384615
2
{"M": 128, "N": 64, "K": 512, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.048415
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.0626
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.055637
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.035662
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.048813
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.036857
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.034368
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.03055
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.02631
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.056855
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.048169
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float32, device='cuda') B = torch.randn(1024, 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}
0.052034
16,777,216
819,200
20.48
4
{"M": 128, "N": 64, "K": 1024, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.048411
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.039344
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.046531
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.057457
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.048658
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.040981
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.052186
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.039744
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.031512
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.023309
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.038373
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 1024) x (1024, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 1024, dtype=torch.float16, device='cuda') B = torch.randn(1024, 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}
0.025653
16,777,216
409,600
40.96
2
{"M": 128, "N": 64, "K": 1024, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.028888
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.046201
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.036986
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.052749
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.040165
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.0385
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.048752
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.054476
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.021841
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.033639
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.038499
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float32, device='cuda') B = torch.randn(2048, 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}
0.022783
33,554,432
1,605,632
20.897959
4
{"M": 128, "N": 64, "K": 2048, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.048344
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.033305
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.031631
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.045646
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.037556
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.049921
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.067096
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.038873
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.035096
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.041919
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.064578
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 2048) x (2048, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 2048, dtype=torch.float16, device='cuda') B = torch.randn(2048, 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}
0.05198
33,554,432
802,816
41.795918
2
{"M": 128, "N": 64, "K": 2048, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.050074
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.072833
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.068474
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.050433
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.069552
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.052773
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.04233
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.053932
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.031757
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.031481
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.04386
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float32, device='cuda') B = torch.randn(4096, 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}
0.046489
67,108,864
3,178,496
21.113402
4
{"M": 128, "N": 64, "K": 4096, "dtype": "float32"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 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}
0.065751
67,108,864
1,589,248
42.226804
2
{"M": 128, "N": 64, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 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}
0.035557
67,108,864
1,589,248
42.226804
2
{"M": 128, "N": 64, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 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}
0.053986
67,108,864
1,589,248
42.226804
2
{"M": 128, "N": 64, "K": 4096, "dtype": "float16"}
import torch def matmul_kernel(A, B): # Matrix multiplication: (128, 4096) x (4096, 64) -> (128, 64) C = torch.matmul(A, B) return C A = torch.randn(128, 4096, dtype=torch.float16, device='cuda') B = torch.randn(4096, 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}
0.057262
67,108,864
1,589,248
42.226804
2
{"M": 128, "N": 64, "K": 4096, "dtype": "float16"}