source stringclasses 13
values | gpu stringclasses 4
values | model stringclasses 13
values | kernel_family stringclasses 7
values | kernel_name stringclasses 328
values | dtype stringclasses 2
values | held_out bool 2
classes | M float64 1 16.4k ⌀ | N float64 8 248k ⌀ | K float64 2.05k 29.6k ⌀ | bs float64 1 1 ⌀ | seq float64 1 16.4k ⌀ | n_heads float64 8 64 ⌀ | head_dim float64 64 256 ⌀ | kv_heads float64 1 64 ⌀ | numel float64 64 531M ⌀ | op_type stringclasses 177
values | gpu_time_duration_ms float64 0 703 | launch_block_size float64 16 512 | launch_grid_size float64 1 1.04M | dram_bytes_sum float64 1.27 94.3B ⌀ | launch_registers_per_thread float64 16 255 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,632 | cast | 0.00298 | 128 | 11 | 20,860 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,632 | cast | 0.0031 | 128 | 11 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.00397 | 128 | 484 | 550,530 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00429 | 128 | 484 | 517,410 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<128, 4, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,072 | other | 0.0112 | 128 | 6 | 543,900 | 46 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.0025 | 128 | 3 | 27,520 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,632 | cast | 0.00304 | 128 | 11 | 21,120 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00317 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.004 | 128 | 507 | 562,820 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00413 | 128 | 507 | 540,450 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 12,288 | other | 0.0096 | 512 | 6 | 559,520 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.00253 | 128 | 3 | 28,030 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00301 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00307 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.004 | 128 | 529 | 575,100 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00416 | 128 | 529 | 564,000 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<256, 2, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | other | 0.0104 | 256 | 6 | 585,630 | 36 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.00253 | 128 | 3 | 28,540 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00298 | 128 | 12 | 21,630 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00323 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.00438 | 128 | 553 | 587,390 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00467 | 128 | 553 | 588,060 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 12,288 | other | 0.00995 | 512 | 6 | 606,620 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.0121 | 128 | 3 | 29,060 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00301 | 128 | 12 | 21,890 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.0031 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.00448 | 128 | 576 | 599,680 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00486 | 128 | 576 | 612,640 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<128, 4, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,072 | other | 0.01293 | 128 | 6 | 638,110 | 46 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.00237 | 128 | 3 | 29,570 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | cast | 0.00307 | 128 | 12 | 22,140 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00314 | 128 | 13 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.0047 | 128 | 601 | 812,670 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00477 | 128 | 601 | 637,730 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 14,336 | other | 0.01021 | 512 | 7 | 655,780 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00237 | 128 | 4 | 30,080 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00301 | 128 | 13 | 22,400 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00314 | 128 | 13 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.00474 | 128 | 625 | 829,060 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.0048 | 128 | 625 | 663,330 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<256, 2, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | other | 0.0111 | 256 | 7 | 683,940 | 36 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00243 | 128 | 4 | 30,590 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00301 | 128 | 13 | 22,660 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.0031 | 128 | 13 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.0047 | 128 | 651 | 845,440 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.0049 | 128 | 651 | 689,440 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 14,336 | other | 0.01021 | 512 | 7 | 706,980 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.0024 | 128 | 4 | 31,100 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00301 | 128 | 13 | 22,910 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00304 | 128 | 13 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.0128 | 128 | 676 | 862,080 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00506 | 128 | 676 | 716,060 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<128, 4, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,584 | other | 0.01184 | 128 | 7 | 740,510 | 46 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.0024 | 128 | 4 | 31,620 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,656 | cast | 0.00301 | 128 | 13 | 23,170 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.0032 | 128 | 14 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | cast | 0.00483 | 128 | 703 | 878,210 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 262,144 | mul | 0.00502 | 128 | 703 | 743,200 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 14,336 | other | 0.01069 | 512 | 7 | 760,220 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.0024 | 128 | 4 | 32,130 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.00304 | 128 | 14 | 23,420 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.0031 | 128 | 14 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.0049 | 128 | 729 | 894,590 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00518 | 128 | 729 | 770,850 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<256, 2, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | other | 0.01165 | 256 | 7 | 790,430 | 36 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.0024 | 128 | 4 | 32,640 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.00304 | 128 | 14 | 23,680 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.00326 | 128 | 14 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00502 | 128 | 757 | 910,980 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00502 | 128 | 757 | 799,010 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 14,336 | other | 0.01082 | 512 | 7 | 815,520 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00243 | 128 | 4 | 33,150 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.00301 | 128 | 14 | 23,940 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.0031 | 128 | 14 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00493 | 128 | 784 | 927,360 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00506 | 128 | 784 | 827,680 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<128, 4, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,584 | other | 0.01357 | 128 | 7 | 851,100 | 46 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00243 | 128 | 4 | 33,660 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,168 | cast | 0.00301 | 128 | 14 | 24,190 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.0031 | 128 | 15 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00499 | 128 | 813 | 943,740 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00512 | 128 | 813 | 856,860 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | other | 0.01123 | 512 | 8 | 872,860 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00246 | 128 | 4 | 34,180 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00304 | 128 | 15 | 24,450 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00307 | 128 | 15 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00515 | 128 | 841 | 960,130 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00512 | 128 | 841 | 886,560 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<256, 2, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | other | 0.01184 | 256 | 8 | 905,120 | 36 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.0025 | 128 | 4 | 34,690 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00307 | 128 | 15 | 24,700 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00307 | 128 | 15 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00506 | 128 | 871 | 976,510 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00518 | 128 | 871 | 916,770 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::func_wrapper_t<float, at::sum_functor<float, float, float>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | other | 0.01123 | 512 | 8 | 932,260 | 32 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,048 | residual | 0.00253 | 128 | 4 | 35,200 | 40 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00298 | 128 | 15 | 24,960 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 7,680 | cast | 0.00307 | 128 | 15 | 26,240 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | cast | 0.00528 | 128 | 900 | 993,920 | 16 |
Qwen3.5-9B_prefill | RTX2080Ti | Qwen3.5-9B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 524,288 | mul | 0.00538 | 128 | 900 | 947,490 | 16 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.