submission 545070 · KernelIndex

Use it

Vendorable · source mirrored · license unknownView source →

No package. Vendor the mirrored source: 28 lines, June 9 Researcher Reciprocity License v1.0.

vectoradd_v2_triton.py

curl "https://kernelindex.com/api/v1/implementations/kernelbot-vectoradd-v2-545070?include=source"

interfacepython

Compatibility

measured onNVIDIA A100

declared hardwareNVIDIA A100

architecturessm_80

dtypesfp16

Benchmark evidence

1 measurement across 1 GPU, fastest first.

Operation / workload

Hardware

Latency

Rank

Observed

FP16 vector additionsuite of 5 cases

NVIDIA A100

896.0µs

#11 of 87

2026-03-13

Run →

Reported · How evidence levels are derived →

Source and license

sourceavailable

revision digestsha256:96278b9c0423d86d6da32553faaef082e63732beda01eb3289c0ac3b2d78128e

license declaredunknown

license concludedunknown

authorsrajesh0042

imported2026-08-15

Kernel source

vectoradd_v2_triton.py28 lines

import os
os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"

import torch
import triton
import triton.language as tl
from task import input_t, output_t

@triton.jit
def vectoradd_kernel(
    a_ptr, b_ptr, out_ptr, n_elements,
    BLOCK_SIZE: tl.constexpr,
):
    pid = tl.program_id(0)
    offsets = pid * BLOCK_SIZE + tl.arange(0, BLOCK_SIZE)
    mask = offsets < n_elements
    a = tl.load(a_ptr + offsets, mask=mask)
    b = tl.load(b_ptr + offsets, mask=mask)
    tl.store(out_ptr + offsets, a + b, mask=mask)

def custom_kernel(data: input_t) -> output_t:
    A, B, output = data
    n = A.numel()
    BLOCK_SIZE = 1024
    grid = ((n + BLOCK_SIZE - 1) // BLOCK_SIZE,)
    vectoradd_kernel[grid](A, B, output, n, BLOCK_SIZE=BLOCK_SIZE)
    return output

Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0

Changes from previous submission

Against this author's previous submission submission 545031.

  ⋯ 1 unchanged lines
  os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"
   
  import torch
+ import triton
+ import triton.language as tl
  from task import input_t, output_t
   
+ @triton.jit
+ def vectoradd_kernel(
+     a_ptr, b_ptr, out_ptr, n_elements,
+     BLOCK_SIZE: tl.constexpr,
+ ):
+     pid = tl.program_id(0)
+     offsets = pid * BLOCK_SIZE + tl.arange(0, BLOCK_SIZE)
+     mask = offsets < n_elements
+     a = tl.load(a_ptr + offsets, mask=mask)
+     b = tl.load(b_ptr + offsets, mask=mask)
+     tl.store(out_ptr + offsets, a + b, mask=mask)
+  
  def custom_kernel(data: input_t) -> output_t:
      A, B, output = data
-     torch.add(A, B, out=output)
+     n = A.numel()
+     BLOCK_SIZE = 1024
+     grid = ((n + BLOCK_SIZE - 1) // BLOCK_SIZE,)
+     vectoradd_kernel[grid](A, B, output, n, BLOCK_SIZE=BLOCK_SIZE)
      return output

scrolls · 28 diff lines total

Best evidence level for this revision: reported

JSON