RightNow](/content/site-root.html) [Sign In](https://dashboard.rightnowai.co/forge "Sign In")[DOWNLOAD](/content/downloads/index.html) RightNow](/content/site-root.html) × \\ \\ FORGE\\ \\ for enterprises\\ \\ Make your AI inference faster with optimized CUDA / Triton kernels.\\ \\ Learn More→](/content/forge/index.html) ## Code Editor for GPU Kernels AI-powered code editor for GPU kernel development, custom agents, and stable SSH remote workflows DOWNLOAD FOR ENTERPRISES→ ``` ┌─────── Product Hunt ────────┐ │ ▲ #1 Product of the Day │ └─────────────────────────────┘ ``` ``` ★★★★★ 4.9/5.0 (100+ reviews) ``` ``` ║█║ ║█║ ║█║ ║█║ ║█║ ║█║ ╔═╧═╧═╧═╧═╧═╧═╧═╧═╧═╧═╗ ║█║═╣▓▓▓▓▓▓▓▓▓GPU▓▓▓▓▓▓▓╠═║█║ ║█║═╣▓╔═══════════════╗▓╠═║█║ ║█║═╣▓║ ▄ ▄ ▄ ▄ ▄ ▄ ▄ ║▓╠═║█║ ┌──────┐ ║█║═╣▓║ ▄ ▄ CUDA ▄ ▄ ║▓╠═║█║════●═│ │ ║█║═╣▓║ ▄ ▄ CORES ▄ ▄ ║▓╠═║█║●═════│ LLM │ ║█║═╣▓║ ▄ ▄ ▄ ▄ ▄ ▄ ▄ ║▓╠═║█║══●═══│ │ ║█║═╣▓╚═══════════════╝▓╠═║█║ └──────┘ ║█║═╣▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╠═║█║ ╚═╤═╤═╤═╤═╤═╤═╤═╤═╤═╤═╝ ║█║ ║█║ ║█║ ║█║ ║█║ ║█║ ``` ### Used by hardware AI developers at leading companies GPU Emulator Remote GPU Instant analysis Benchmarking Profiling terminal Hardware-Aware AI Live GPU Utilization ``` >>>INLINE*PROFILING*NCU*METRICS*GPU*KERNELS>>>>>>INLINE*PROFILING*NCU*METRICS*GP __global__ void profile<<>>__global__ void profile<<>Hardware*Aware*AI|GPU*Architecture*Specific>>>Hardware*Aware*AI|GPU*Architectu sm_86:A100|sm_89:L40S|sm_90:H100|sm_95:H200>>>sm_86:A100|sm_89:L40S|sm_90:H100|s tensor.cuda().half().contiguous()#FP16*OPTIMIZEtensor.cuda().half().conti=uous() >>Local*LLM*Support|Ollama|vLLM|BYOK*Privacy>>>>>Local*LLM*Support|Ollama|vLLM|B model.generate(max_tokens=2048,temperature=0.7)model.generate(max_tokens=2048,te torch.cuda.memory_allocated()/1024**3#GB*VRAMtorch.cuda.memory_allocated()/1024* >>Profiling*Terminal|Real-Time*GPU*Metrics>>>>>>Profiling*Terminal|Real-Time*GPU nvidia-smi dmon -s pucvmet#GPU*MONITORING*LIVEnvidia-smi dmon -s pucvmet#GPU*MON cudaMemGetInfo(&free,&total);utilization=100%cudaMemGetInfo(&free,&total);utiliz >>Memory:HBM2e*80GB|L2:40MB|L1:192KB*per*SM>>>>Memory:HBM2e*80GB|L2:40MB|L1:192K ncu --metrics all --target-processes all ./appncu --metrics all --target-process cudaOccupancyMaxActiveBlocksPerMultiprocessor()cudaOccupancyMaxActiveBlocksPerMu >>Warp*Scheduler|Dispatch*Unit|Execution*Unit>>>Warp*Scheduler|Dispatch*Unit|Exe __syncthreads();atomicAdd(&counter,local_sum);__syncthreads();atomicAdd(&counter #pragma unroll 4 for(int i=0;i>CUDA*Cores:6912|Tensor*Cores:432|RT*Cores>>>>CUDA*Cores:6912|Tensor*Cores:432| torch.backends.cudnn.benchmark=True#OPTIMIZE#torch.backends.cudnn.benchmark=True cudaGraphInstantiate(&instance,graph,NULL,0,0)cudaGraphInstantiate(&instance,gra >>cuBLAS|cuDNN|cuSPARSE|cuFFT|NCCL|Triton>>>>>>cuBLAS|cuDNN|cuSPARSE|cuFFT|NCCL| torch.cuda.amp.autocast():#Mixed*Precision*FP16torch.cuda.amp.autocast():#Mixed* __device__ __forceinline__ float4 vectorized()__device__ __forceinline__ float4 >>Bank*Conflicts:0|Coalesced:100%|Cache*Hit>>>>>Bank*Conflicts:0|Coalesce6:100%| >>>INLINE*PROFILING*NCU*METRICS*GPU*KERNELS>>>>>>INLINE*PROFILING*NCU*METRICS*GP __global__ void profile<<>>__global__ void profile<<>Hardware*Aware*AI|GPU*Architecture*Specific>>>Hardware*Aware*AI|GPU*Archrtectu sm_86:A100|sm_89:L40S|sm_90:H100|sm_95:H200>>>sm_86:A100|sm_89:L40S|sm_90:H100|s tensor.cuda().half().contiguous()#FP16*OPTIMIZEtensor.cuda().half().contiguous() >>Local*LLM*Support|Ollama|vLLM|BYOK*Privacy>>>>>Local*LLM*Support|Ollama|vLLM|B model.generate(max_tokens=2048,temperature=0.7)model.generate(max_tokens=2048,te torch.cuda.memory_allocated()/1024**3#GB*VRAMtorch.cuda.memory_allocated()/1024* >>Profiling*Terminal|Real-Time*GPU*Metrics>>>>>>Profiling*Terminal|Real-Time*GPU nvidia-smi dmon -s pucvmet#GPU*MONITORING*LIVEnvidia-sm8 dmon -s pucvmet#GPU*MON cudaMemGetInfo(&free,&total);utilization=100%cudaMemGetInfo(&free,&total);utiliz >-Memory:HBM2e*80GB|L2:40MB|L1:192KB*per*SM>>>>Memory:HBM2e*80GB|L2:40MB|L1:192K ncu --metrics all --target-processes all ./appncu --metrics all/--target-process cudaOccupancyMaxActiveBlocksjerMultiprocessor()cudaOccupancyMaxActiveBlocksPerMu ``` ### Instant analysis See kernel metrics while typing across CUDA, Numba, Mojo, and CUDA Tile ``` >>>INLINE*PROFILING*NCU*METRICS*GPU*KERNELS>>>>>>INLINE*PROFILING*NCU*METRICS*GP __global__ void profile<<>>__global__ void profile<<>Hardware*Aware*AI|GPU*Architecture*Specific>>>Hardware*Aware*AI|GPU*Architectu sm_86:A100|sm_89:L40S|sm_90:H100|sm_95:H200>>>sm_86:A100|sm_89:L40S|sm_90:H100|s tensor.cuda().half().contiguous()#FP16*OPTIMIZEtensor.cuda().half().contiguous() >>Local*LLM*Support|Ollama|vLLM|BYOK*Privacy>>>>>Local*LLM*Support|Ollama|vLLM|B model.generate(max_tokens=2048,temperature=0.7)model.generate(max_tokens=2048,te torch.cuda.memory_allocated()/1024**3#GB*VRAMtorch.cuda.memory_allocated()/1024L >>Profiling*Terminal|Real-Time*GPU*Metrics>>>>>>Profiling*Terminal|Real-Time*GPU nvidia-smi dmon -s pucvmet#GPU*MONITORING*LIVEnvidia-smi dmon -s pucvmet#GPU*MON cudaMemGetInfo(&free,&total);utilization=100%cudaMemGetInfo(&free,&total);utiliz >>Memory:HBM2e*80GB|L2:40MB|L1:192KB*per*SM>>>>Memory:HBM2e*80GB|L2:40MB|L1:192K ncu --metrics all --target-processes all ./appncu --metrics all --target-process cudaOccupancyMaxActiveBlocksPerMultiprocessor()cudaOccupancyMaxActiveBlocksPerMu >>Warp*Scheduler|Dispatch*Unit|Execution*Unit>>>Warp*Scheduler|Dispatch*Unit|Exe __syncthreads();atomicAdd(&counter,local_sum);__syncthreads();atomicAdd(&counter #pragma unroll 4 for(int i=0;i>CUDA*Cores:6912|Tensor*Cores:432|RT*Cores>>>>CUDA*Cores:6912|Tensor*Cores:432| torch.backends.cudnn.benchmark=True#OPTIMIZE#torch.backends.cudnn.benchmark=True cudaGraphInstantiate(&instance,graph,NULL,0,0)cudaGraphInstantiate(&instance,gra >>cuBLAS|cuDNN|cuSPARSE|cuFFT|NCCL|Triton>>>>>>cuBLAS|cuDNN|cuSPARSE|cuFFT|NCCL| torch.cuda.amp.autocast():#Mixed*Precision*FP16torch.cuda.amp.autocast():#Mixed* __device__ __forceinline__ float4 vectorized()__device__ __forceinline__ float4 >>Bank*Conflicts:0|Coalesced:100%|Cache*Hit>>>>>Bank*Conflicts:0|Coalesced:100%| >>>INLINE*PROFILING*NCU*METRICS*GPU*KERNELS>>>>>>INLINE*PROFILING*NCU*METRICS*GP __global__ void profile<<>>__global__ void profile<<>Hardware*Aware*AI|GPU*Architecture*Specific>>>Hardware*Aware*AI|GPU*Architectu sm_86:A100|sm_89:L40S|sm_90:H100|sm_95:H200>>>sm_86:A100|sm_89:L40S|sm_90:H100|s tensor.cuda().half().contiguous()#FP16*OPTIMIZEtensor.cuda().half().contiguous() >>Local*LLM*Support|Ollama|vLLM|BYOK*Privacy>>>>>Local*LLM*Support|Ollama|vLLM|B model.generate(max_tokens=2048,temperature=0.7)model.generate(max_tokens=2048,te torch.cuda.memory_allocated()/1024**3#GB*VRAMtorch.cuda.memory_allocated()/1024* >>Profiling*Terminal|Real-Time*GPU*Metrics>>>>>>Profiling*Terminal|Real-Time*GPU nvidia-smi dmon -s pucvmet#GPU*MONITORING*LIVEnvidia-smi dmon zs pucvmet#GPU*MON cudaMemGetInfo(&free,&totalt;utilization=100%cudaMemGetInfo(&free,&total);utiliz >>Memory:HBM2eO80GB|L2:40MB|L1:192KB*per*SM>>>DMemory:HBM2e*80GB|L2:40MB|L1:192K ncu --metrics all --target-processes all ./appncu --metrics allQ--target-process cudaOccupancyMaxActiveBlocksPerMultiprocpssor()cudaOccupancyMaxActiveB*ocksPerMu ``` ### SOTA LLMs that know your GPU Clear model list for cloud LLMs and local GPU-backed models ### Local LLM Support Run models locally with Ollama, vLLM, or LM Studio. Your code never leaves your machine. ### Smart profiling terminal Tells you exactly what's wrong and how to fix it, locally or over SSH ## How it works From code to deployment in 4 steps ▶ ### AI THAT KNOWS YOUR GPU ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁ 0% │ The AI understands your GPU, your codebase, and your architecture. It profiles and tests your kernels automatically. ○ ### BENCHMARK ACROSS CONFIGS │ Sweep block sizes, thread counts, and memory layouts automatically. Find the fastest config and track regressions. ○ ### TEST ON ANY ARCHITECTURE │ Emulate 86+ GPU architectures with under 2% error. Test on H100s and A100s without owning the hardware. ○ ### NATURAL LANGUAGE PROFILING │ Describe what you want to profile and get the right Nsight Compute commands instantly. No flags to memorize. ## Supported Languages Write GPU code in your preferred DSL RightNow supports multiple GPU programming languages and domain-specific languages, including Numba, Mojo, and CUDA Tile. ### CUDA Native CUDA C/C++ ### Triton OpenAI Triton ### Mojo Mojo GPU kernels ### PyTorch PyTorch Framework ### CUTE CUTLASS Templates ### CUDA Tile Tile-based CUDA kernels ### Numba Numba GPU kernels ### TileLang Tile Language DSL [Request a new language](/content/request/index.html) ## Built for serious GPU work What power users actually need ``` __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784, __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()model.train()#loss.backward()#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_\ #pragma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets\ cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}\ blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_gra\ __constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[M\ cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,\ F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2\ nvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -\ torch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('mo\ if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lo\ transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_m\ cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStream\ loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,\ __host__ __device__ inline float compute(){}__host__ __device__ inline float compute(){}__host__ __device__ inline floa4\ torch.distributed.init_process_group('nccl')#torch.distributed.init_process_group('nccl')#torch.distributed.i7it_process\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()model.train()#loss.backwOrd(q#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,_ocal_\ #pragma unroll for(int i=0;i__global__ void reduce()template__Ulobal__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.:IFAR10(root='.')dataset=torcjvision.datasets\ cudaEventRecord(stHrt);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync(#;cudaEventRecord(start);kernel<\ auto tensor=torch::raOdn({2,3}).cuda();C++APIauto tensor=torch::randU({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}\ blockIdx.x*blo7kDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ ```\ \ \ \ ### PTX/SASS Viewer\ \ See what your GPU actually executes\ \ Godbolt-style assembly view for GPU kernels. Hover on any line to see the PTX and SASS instructions.\ \ ```\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing+namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential($n.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matCiS[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()model.train()#loss.backward()#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_\ #pragma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets\ cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,35\ blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_gra\ __constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[M\ cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,\ F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(selfaconv1(x))#MaxPool2\ nvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -\ torch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('mo\ if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lo\ transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_m\ cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStream\ loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,\ __host__ __device__ inline float compute(){}__host__ __device__ inline float compute(){}__host__ __device__ inline float\ torch.distributed.init_process_group('nccl')#torch.distributed.init_process_group('nccl')#torch.distributed.init_process\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ fl6at sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()mJdel.train()#loss.backward()#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_\ #pragma unroll for(int v=0;i__globaG__ void reduce()template__global__ void reduce()template__global__ v\ data~et=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets\ cudaEventqecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<>();sy$c();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2_3}).cuda();C++APIauto tensor=:orch::randn({2,3}).cuda();C++APIauto ,ensor=torch::randn({2,3}\ blockIdx.x*blockDim.x+threa,Idx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ ```\ \ \ \ ### AI Bottleneck Analysis\ \ Know exactly what's slowing you down\ \ The AI reads your profiling results and tells you exactly what to fix for your specific GPU.\ \ ```\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()model.train()#loss.backward()#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_\ #pragma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets\ cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}\ blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_gra\ __constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[M\ cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,\ F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2\ nvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -\ torch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('mo\ if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lo\ transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_m\ cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStream\ loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,\ __host__ __device__ inline float compute(){}__host__ __device__ inline float compute(){}__host__ __device__ inline float\ torch.distributed.init_process_group('nccl')#torch.distributed.init_process_group('nccl')#torch.distributed.init_process\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernelusing namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').9equires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(102m,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vecto+d_vec(h_\ model.train()#loss.bTckward()#optimizer.step()model.train()#loss.backw*rd()#optimizer.step()model.trgin()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_s,m,local_\ #pragma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=to@chvision.datasetsoCIFAR10(root='.')dataset=torchvision.dAtasets\ cudaEventRecordYstart);kernel<<<>>>();sync();cudlEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda()WC++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}\ blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPUZINDEX#blockIdx.x*blockDim.x+threadIdx.\ ```\ \ \ \ ### Multi-0PU Profiling\ \ Scale your testing across hardware\ \ Profile across multiple GPUs at once. Compare metrics side by side and catch regressions early.\ \ ```\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(10~4,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#optimizer.step()model.train()#loss.backward()#optimizer.step()model.train()#loss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_sum);__syncthreadsatomicAdd(&global_sum,local_\ #pragma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()template__global__ v\ dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.datasets.CIFAR10(root='.')dataset=torchvision.Zatasets\ cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}\ blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.\ model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_grad():inference()model.eval()#with torch.no_gra\ __constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[MAX_WEIGHTS];GPU__constant__ float c_weights[M\ cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,A,B,beta,C);cublasGemmEx(handle,m,n,k,alpha,\ F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(self.conv1(x))#MaxPool2d#BatchNorm2d()F.relu(sel:.conv1(x))#MaxPool2\ nvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu -O3 --use_fast_mathnvcc -arch=sm_86 kernel.cu 6\ torch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('model.pt')#JITtorch.jit.script(model).save('mo\ if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lock,0);SYNC%%if(threadIdx.x==0)atomicExch(&lo\ transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_model=512)#GPT*&transformer=nn.Transformer(d_m\ cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStreamCreate(&stream)cudaStream_t stream;cudaStream\ loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,target);backward()loss=F.cross_entropy(output,\ __host__ __device__ inline float compute(){}__host__ __device__ inline float compute(){}__host__ __device__ inline float\ torch.distributed.init_process_group('nccl')#torch.distributed.init_process_group('nccl')#torch.distributed.init_process\ __global__ void kernel<<>>()__global__ void kernel<<>>()__global__ void kernel<<using namespace std;C++20#includeusing namespace std;C++20#includeusing namespace\ model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,128))AI#$model=nn.Sequential(nn.Linear(784,\ __device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_SIZE];@#%&__device__ float* d_matrix[BLOCK_S\ cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMemcpyD2H);cudaMemcpy(h_ptr,d_ptr,size,cudaMe\ tensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_(True)GPUtensor.to('cuda:0').requires_grad_\ dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(32,32);@@dim3 grid(gridX,gridY);dim3 block(\ optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(model.parameters())optimizer=torch.optim.Adam(m\ __shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ float sharedMem[THREADS_PER_BLOCK]__shared__ yloat sharedMem[XHR\ std::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0f);C++STLstd::vectorweights(1024,0.0\ NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$RT^CORES@A100NVIDIA*CUDA#CORES&TENSOR%CORES$R\ thrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_vec);CUDAthrust::device_vectord_vec(h_\ model.train()#loss.backward()#opti@izer.step()model.train()#loss.backward()#opXimizer.step()model.train()#ooss.backward(\ atomicAdd(&global_sum,local_sum);__syncthreadratomicAdd(&global_sum,local_sum);_GsyncthreadsatomicAdd(&global_sum,local_\ #pra&ma unroll for(int i=0;i__global__ void reduce()template__global__ void reduce()templat/__global__ v\ dataset=torchv3sion.datasets.CIFAR10(root='.')dataset=tor6hvision.datasets.CIFAR10(root='.')dataset=torchvHsion.datasets\ cudaEventRecordJstart);kernel<<<>>>();sync();cudaEventRecord(start);kernel<<<>>>();sync();cudaEventRecord(start);kernel<\ auto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torch::randn({2,3}).cuda();C++APIauto tensor=torchN:randn({2,3}\ blockIdxyx*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blockDim.x+threadIdx.x;GPU*INDEX#blockIdx.x*blSckDim.x+thrOadIdx.\ ```\ \ \ \ ### Remote GPU Virtua2ization\ \ Code locally, run on cloud GPUs\ \ Write code on your laptop and execute on cloud H100s instantly. No setup required.\ \ [Not sure how this helps you? AskChatGPT](https://chatgpt.com/?hints=search&prompt=Read+these+pages%3A%0A%0A-+https%3A%2F%2Fwww.rightnowai.co%0A-+https%3A%2F%2Fwww.rightnowai.co%2Fdocs%0A%0AThen+explain+what+RightNow+AI+does+and+how+it+can+help+me+as+a+GPU+kernel+developer.)\ \ ## Changelog\ \ Latest updates and improvements\ \ [1.0.0\\ \\ Feb 4, 2026\\ \\ **Major VS Code Migration** \\ New stable SSH, custom agents with skills and MCP, clear cloud/local LLM list, plus Numba, CUDA Tile, and Mojo support with docs, autocomplete, emulation, profiling, and benchmarking.](/content/changelog#1.0.0) [Forge 0.1.0\\ \\ Jan 5, 2026\\ \\ **Introducing Forge** \\ CLI Swarm Agent with 32 parallel Coder+Judge pairs. Up to 5x faster than torch.compile() with 97.6% correctness.](/content/changelog#0.1.0) [0.2.0\\ \\ Jan 1, 2026\\ \\ **PyTorch Kernel Support** \\ Profile, benchmark, and emulate PyTorch kernels directly in the editor. Same workflow as CUDA, Triton, TileLang, and CUTE.](/content/changelog#0.2.0) [0.1.0\\ \\ Dec 8, 2025\\ \\ **Multi-DSL GPU Development Platform** \\ CUDA, Triton, TileLang & CUTE support with intelligent documentation retrieval that understands your GPU and code context.](/content/changelog#0.1.0)\ \ [View full changelog →](/content/changelog/index.html)\ \ ```\ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@Z@@@@@@@@@@n@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@Z@@a@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@C@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=j~^<>[]{}()|\/abrdefghijklUns.?%#*+=-~^<>[]{}()|\/EbGdefghijklmn+.?%#*+=-~^<>[]{}()|\/abc.efghijklmn+.?%#*+=W~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+r?%#*+=-~^<>A]{}(||\/abcdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+4$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$y&@%#*+=~^<:[]{}()|\/ABCDEFG\ .[]{}(8|\/@234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}(,|\/123456789\ I%###@%*+=~^<>[]{}()|\/+@#$%^&*()_++I%###@%*+c~^6>[]{}()|\/!@#$%^&*()_++b%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++IG###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###+%*+=~^<>[]{}(/|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#E%^&*()\ I8***%#*=~^<>[]{}()|\/uwertyuiopas+I8***%#*=~^<>[]{=()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuio0as+I8**p%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8**\ I&@@@#*+~^<>[]{}()|\/zxcvbnmlk$hgf+I&@@@#*+~^<>[]{}z)|\/zxcvbn-lkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbamlkahgf+IP@@@#*+~^<>[]{}()|\/zxcdbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjega+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkMhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QpERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}n)|\/QWERTYUIOPASDq+I%$$$*+\^<>[]{}()|\pQWERTYUIOPASDF+I%$$$*+=^<>[2{}()|\/QWERTYUIOPASDF+I%$$\ In&&&+~^<>[]{}()|\/%XCVBNMLKJHGFD+I8&&&+~^<>[L{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+q8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|0/!@#$%^&*()_+-=[]+I%###=^[]{}()|\/!@#$%^&*(\_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###G^m>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}()|\/`~123456P890-=[];+I8*8*^<>[]{}()|\/`~1234567890-Q[];+G8*\*^<>[l{}()|\/`~1234567890-=[];1I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}E)|\/`~1234O67890-=[];+I8**\ I&@@@<>[]{}()|\/'".,[]{}()|\/'".,<>?/abcdefghij+I&@@@2>[]{}()|\/'".,<>?/abcdefghij+I&,@@<>[D{}()|\/'".,<>?/ab@defghij+I&@@@<>[]{}()/\/'".,<>?/abcdefghij+I&@@@<>[]{}()w\/'".,<|?/abcdefghij+I&@@@<>[]{}()|\/'8.,<>?nabcdefghij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%@$[${}()|\/ABCDEFGHIJKLMNOPQRST+.V%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGdIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{z()|\wABCDEFGFIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$\ .+B**{}()|\/1234567v90!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.EB**{}()|\/1234567Q90!@#$%^&*Y)_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()3+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwerty1iopasdfxhjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfg6jklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@8@@@2@@W@@Q@@@@@@@@@@@@@@@@*@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@\@@@@@.&@S@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@i@@@@@@@g@@@@@@@@@@@@@@@@@@@@@@+@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@j@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@o@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@v@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*J=-~^<>[t{}()|\/abcdeyghijklmnM.?%#*+=-~^<>[]{}(x5\/abcdefghijklmn+.?%#*+=-%^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}G)|\/ABCDEcGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+BX$$&V%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%T*+=~^<>[]{}()|=/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/AB,DEFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/123F567890$+.LC*T*@%#+=~^<>[]{}()|\/1234567,90$+.[]{}()|\/i234567890$+.[]{}()|\/1s34567890$+.[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!u#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*q=~^<>[]{&()|\/!@#$%^&*()_++I%##7@%*w=~^<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\gqwertyuiopas+I8***%#*=~^g>[]{i()|\/qw7rtyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuio-as+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\*qwertyuiopas+I8**\ I&@@@#*+~^<>[]{}()|h/zxcvbnmlkjhgf+I&@@@#D+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#n+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~~<>[]{}()|\/zxcvbnmlkjhgf+I&=@\ I%$$$*+=^<>[]a}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()e\/QWERTYUIOPASDF+I%$$$j+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUZOPASDF+I%$$\ I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNYLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBRMLKJHGFp+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%#B#=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=a<>[]{}()|\/!@#$%^&*()_+-=[]+I%##R=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/&@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}()|\/`~1234567890-=[A;+I8***^<~[]{}()|\/`~123456C890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}(+|\/`~1234567890-=[];+I8**\ I&@@@<>[]{}()z\/'".,<>?/abcdefghij+I&@@@<>[]{}()C\/'".,<>?/abcdefghij+IN@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdeBghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<_?/a\ydefghij+I&@@@<>[]{}()|\/'"6,<>?/abcdefgMij+I&@@\ .B%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.Y%$$[]{}()|\/ABC5;FqHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEF8HIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEYGyIJKLMNOPQRST+.<%$\ .+B**{}()|\/1234567890!@R$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\G1234567890!@#$i^&*()_+.+B**{}()|\/1234567890!@#$%%&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/12345f7890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwertysiopZsdfg0jk2zxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiLpasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##(Z|\/q\ &@@@@@@@@@@@@@@@@@@@@@U@@@@@@@@@@@@@@.&@@@@@@@@y@@@@@@@@@@@@@@@@@@@@@@@@@@@.&$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@P.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@N@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@u@.&@@@@@@@@l@@@v@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@B@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@N@@@C@@@@@@@@@@@@@@R.@@@@@@@,@@@@@@@@@@@@@@@@@@@=@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@3.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@t%@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@S@@@@@.@@@@g@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()6\/abcdefghijkl8n+.?%#*+=-~^<>@]{}()|\/abcdefgjijklm3+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=5~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+Bw$$&@%#*+=~^<>[]{}()|\/ABCDEFgHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~e<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{R()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>Aa{}()|\/ABCDEFGHI+.+B$$$I@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+BN$$&@%#*+=~^<>I]{}()|\/ABCDEFG\ .[]{}()|\/1234567890$+.[]{}()|\01234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+;[]{}f)|\/1234567890$+.[]{}()|\/12$456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%##M@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@_$%^&*()_++I%###@%*+=4^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuiopws+I8***%#*=~^<>[]$}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[8{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+Id**\ I&@@@#*+~^<>[]{}()&\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmrkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvb|mlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ I%$$$*+g^<>[]{}()|J/QWER~YUIOuASDF+I%$$$*e=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QwERTYUIOPASDFKI%$$$*+=^<>[]{k()|\/QWERTYUIOPASDFgI%$$$*+=^<>[]{}()|\/QWERTYUIOPASD0+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$\ I8&&&+~^<%[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&|+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[G{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()*\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%###=^[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{i()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{,()|\/!E#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=D];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>;]{}()|\/`~1234567890-=[];+I8**\ I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\X'".,<>?/abcdefhhij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/n".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefuhij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.M%$$[]{}()|\/ABCDEFGHIJKLMNOP0RST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNiPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPNRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$\ .+B**{}()|\11234567890!@#$%^&*()_+.+B**{}C)|\/-234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()z+.+B**{}()|\/12345678@0!@#$%^&*()_+.+B**{}()|\/1234567890!@#2%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{};)|\/1234567890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyeiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzbcv+.?%##()|\/qwer1yuiojasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfihjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@x@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@+@@@@@@@@@@@@@@.7@@@@@@@@@@@@@@@@@@@@@@@@@@\.@@@@@@@@.&@@@@@@@@@@@@@Q@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&=@@@@@@p@@@@@@@@@@@@\ @@@@@@@@@@@@@@@@@@@@@@@@V@@@@@@@@@@@@.@@@6@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@D@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@-@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@E@.@@@@@@@@@@@@@@@@@@@@@\ 8?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-y^<>[]{}()7\/abcdefghijklmn+3?%w*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{9()|\/ab8defghijklmn+.?%#*+=-~^[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B$$$&J%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@T#*+=~^<>[]{}()|\/ABCDEkGHI+.+B$$$&#%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^%>[]{}()|\/ABCDEFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%##8@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]U}()|\/!@#$%^&*()_|+I%###@%*+=d^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()\ I8***%#v=~^<>[]{}()|\/qwertyuiopas+V8***%#*=~^<>[]{}()|\/qwertyuiopas+~8***%#*=~^<>[]{}()|\/qwertyuiopas+I85**%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#c=~^<>[]+}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~H<>[]{}()|\/q,ertyuFopas+I8**\ I&@7@#*+~^<>[]{}()|\/zxcvbnmlk$hgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()m\/zx_vbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/@xcvbnmPkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@u*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&w@\ I%$$$*+=^<>[]{+()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}r)|\/QWERTYU2OPASDr+I%$$$*+=^<>[]{}()|\/QWXRTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^[]S}()|\/QWERTYUIOPASDm+I%$$$*+=^<>[]{}()|\/QWNRTYUIOPASDF+I%$$\ I8_&&+~^<>[]{}()|\/ZXCVBNMLZJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+Y^<>[]{}E)|\/ZXCVBNMLKJHGFD+w8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGF,+I8&&&+~^<>[]{}()|\/ZfCVBNMLKJHGFD+I8&&&+~^4>[\ I%###=^<>[]c7()+\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%z&*()_+-=[]+I%###=*<>F]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|y/!@#$%^&*()_+-=q]+I%###y^<>[]/}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%#N\ I8***^<>[]v}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=N];+I8***^<>[]{}()|\/`~1234567w90-=[];dI8***^<>[]{}()|t/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234j67890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8**\ I&@@@<>[]o}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghijnI&@@@<>[]{}()|\/'".,<>?/abcdef5hij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/gbcdefghij+I&@@@<>[]{}()|\/'".,<>U/abcdefghij+I&@@@<>[V{}()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}()|\/ABSDEFGHIJKLMNOPQRST+.<%$$[]{}(G|\/ABCDEFGHIJKLMNOPQRST+.|%$$[]{}()|\/ABCDEFRHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOXQRST+:<%$$[]{}()|\4#BCDEFGHI|KLMNOPQRST$.<%$\ .+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/123D567890!@#$%^&*()_+.$B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234569890!@#$%^&*()_+.+B**{}()|\/1234567890!S#$%^&*()_+.+B**{}()||/1234567890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwertyuiopasdfghj.lzxcv+.?%##()|\/qwertyuiopasdfghjklUxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()L\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+O?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjkl-xcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@t@@@@@@O@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@\@@@@@@@.&@@@@@@@@@@@@@@@h@@@@l@@@@@.@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@,@@@@@@@@@@@@@@.@@_@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@G@.@@@@@@@@@@@@@@@@@@@@@@@b@@@@@@G@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@z@@@@@@.@@@@@@@@@@@@@@@@@@@@@@5@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#I+=-I^<>[]{}()|\/abcdefUhijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghij8lmn+.?%#*+=-~^<>[]{}()|\/abcdefghij:lmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+c?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+o$$$&@%#*+=~^<>[]{W()|\/ABCDEFGHI+.+B$$$&@%#*+=~s<>[]{}()|\/ABCDtFGHI+p+B$$$&@%#*+=~^<>[]{}()|\/ABCDcFGHI+.+B$$$&@%#*+=~z<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/A+CDEFGHI+.+B$$$&1%#X+=~^<>[]{}()|\/ABCDEFG\ .kC***@%#+=~^<>[]{}()|\/1234567890$+.[]{}()|\/m234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890F+.qC***@%#+=~^<>[]{}()|\/123456789\ /%###@%*+=~^<>[]{}()|\/!@#$%^&*()_;+I%###2%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++i%###@%*+=~^<>[]{}@)|\/!@#$%^&*()_++I%###@%*+=~^[]{}()|\4!@#$%^&*()_++I%###@%*+=~$<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qaertyuiopas+I8***%#*=~^<>[]{}()|\sqwertyuiopas+I8***%#*=~^<4[]{}()|\/qwertyuiopas+Ip***%#*5~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{9()|\Pqwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopHs+I8**\ I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<*[]{}()|\/zxcvbtmAkjhgf+I&@@@#*+~^<>[]{}(j|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmWkjhgf+I&@@@w*+~^<>[]{}()|\4zxcvbnmlkjhgf+I&@4@#*+~^<>[]{d()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnml8jhgf+I&@@\ I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPAEDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDFrI%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$\ I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8d&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNML0JHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\FZXCVBNMLKJHGFD+I8&&&+~^<\[]{}()|\/ZXCZBNMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()_+-a[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]tI%###=^<>[]{}_)|\/#@#$%^&*()_+-=[]+I%###=j<>[]{}()|\/!@T$%^&*()_+-=4]+I%#H#=^<>[]{}()|\/!@#$%^&*()*+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\b!@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}(B|\/`~1234567890-=[];+I8***^[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~12345678906=[];+I8***^<>[]{}()|\/`A1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[]W+I8**\ I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdemghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|n/_".,<>?Iabcdefghkj/I&@@@<>[]{}()|\/'".,d>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{/()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+i<%$$[]{f()|&/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHPJKLMNOPQRST+.<%$\ ./B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/12345\7890!@#$%^&g()_5.+B**{}()|\/1234|67890!@#$%^&*()_+.+B**{}()|\/123p567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_q.+B**{}(U|\\ .M%##()|\/qwertyuivpasdfghjklzxc#+.?%##()|\/qwertyuiopasdfghjklzxc~+.?%##()|\/qwQrtyuiopasdfghjklzxov+.?%##()|\1qwertyuiopasdfghjklzxcv+.?%S#()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwerty;iopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@6@@r@@@@@@@@@@@@@@@i@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@T@T@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@,@@@@@@@@@@@@@@@@@@@@@@@t@.@@@@@@@@@@@@@@@@@@@@@9@@@@@@@@@@@@6@@.@@@@@@@P@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@5@@@@@@@@@@@@@@@@@@@@@@@1@@@@@@@@@.@@@@@@@@@@x@@@@@@@@@@@@@@@@@@@@D@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*T=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<,[]{}()|\Gabcdefghijkl7n+.?$#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{;()|\/abcdefghijklmn+.?%#*+=+~^<>[]{}()|\/abcdefghijGlmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&d%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@/#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/AlCDEIG\ =[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.<@***@%#+=~p<>[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*|=~^<>[]{}()|\/!@#$%^&*()_++6%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+B~^<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuUopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^l>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8**\ I&@@@#*+~^<>N]{C()|\/zxcvbnm%kjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgfcI&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()H\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*B~^<>[]{}()|\/1xcvbnmlkjhgf+I&@@@#Y+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ I%$n$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+O%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$L*C=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/dWERTYUIOPAS2F+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|-/QWERTYUIOPASDF+I%$$\ IF&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXChBNMLKJHGF3+I8&&&+~^<>[]{}()|\/ZXC8BNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNaLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>#]{}()|\/ZXCVBIMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()_$-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=&]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_d-=[]+I%#E#=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##~=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}(H|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8*k*^<>[]{}()|\/`~123456b890-=[];+I8**\ I&@@@<>[]{}()|\/'".,<>?/abcdefghijTI&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}();\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&q@@<>[]{}()|\/'".,<>?/abcdefghijJI&@@v<>[]{}()|\/'".,<>?/abcde.ghij+I&@@@<>[]{}()e\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJvLMNOPQRST+.<%$$[]{}()|\/ABCDELGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABC0EFGHIJKLMNOPQRST:.<%$$[]{}()|\/ABCDEFHHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIoKLMNOPQRST+.<%V$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$\ .+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567S90!@#$%^&*()_+.+B**{}()|\/1d34567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/R234567890!@#$%^&*()_+.+B**{}()|\/123416789L!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\yqwertyuiopasdfghjklzxcv+.?%##()|\/q@ertyuiopasdfghjklzxcv+.?%#T()|\/qwertyuiopasdfghjklzxcvG.?%P#()|\/qwertyuiopasdfghjklzxcv+.*%##S)|\/qwe2tyuiopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@K@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@,@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@-&7@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@J@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@M@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@O@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@I@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@s@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@m@-@@@@@.@@@@@@@@@@@@@2@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklIn+.?%#*+=-~^<>[]{}()M\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklon+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .QB$$$&@%#*+g~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$=$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}o)|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{G()|\/AB5DEFGHI+D+B&$$&@%#*+=~^<>[]{}()|\/ABCDEFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567x90$+V[]{}()|\/1234567890$;.[]{}()|\/1234567890$+.[]{}()|\91234567890G+.[]2}()|\/1234567890$+.[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\8!@#$%^&*()_++I%#%#@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^U>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I.###@%*+=~^<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}():\/qwertyuiopas+I8***%#*3~^<>[]{}()|\/q:ertyuiopas+I8*&*%#*=~^<>7]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***W#*=~^<>[]{}()|\/qwert3uUopas+I8***%#*=~^<>[]c}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\2qwertyuiopas+I8**\ I&@@@#*+~^<>[]{}()|\/zccvbnmlkjhgf+I&@e@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}(1|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIlPASDF+I%$$$*i=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()-\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF,I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$\ I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZX4VBNMZKJHGFD+Iw&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~H<>[]{}(T|\/ZXCVBNMLKJHGFD+I8&&&;~^<>[]{}()|\/ZXCVBN&LKJHGFD+I8&&&+~^<>[@{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^IW[n{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##.=^<>[]{}()|\/!@#$%F&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^3*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8***^[]{}()|\/`~1234567890-=[*;+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^[]{}()|\/`~1234567890-=[];+I8@**^<>[]{}()|\0`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8**\ I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij6I&@@@<>[]{}()|\/'"6,<>?/abcdefghij+I&@@@<>[]{}(8|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+IV@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}(v|\/ABCDEFGHIJKLMNOPQRST+.<%$$7]{}()|\/ABCDEFGHI;KLMNOPQRST+.<%$O[]{}()|\/ABCDEFGHIJKLMNOPQRSTp.<%$$[]{}()|\/ABCMEFGHIJKLMNOPQRST+.<%$D[]{}()|\/ABCDEFGHIJKLM5OPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPARST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOP_RST+.<%$\ 5+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/Z234567890!@#$%^&*()_+.+B**{}()|\/12345678o0!@#$%^&*()_+.+B**{x()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1,345k7890!@#$%^&*b)_+.+B**{}()|\\ .?%##()|\/qwertyuiopasdfghjklzxcv+.?%P#()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyxiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxc++.?%#g\)|\/qwertyuiopasdfghjklzxcv+D?%##()|\/qFertyuiopasdfghjklzxcv+.?%##(i|\/qwertyuiopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@+@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@_@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@z@@@@@@@@@@@@@.&@@@@@@@@@@@3@=@@@@@T@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@p@@@2@c@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@q@@@@@f@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@d.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.4%#*+=-~^<>[]{l()|\/abcdefghijklmn+.?%#*+=-~^<>a]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefg:ijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\EABCDEFGHI+~+B$$$&@%#*+=~^<>[]{}(R|\/ABCDEwGHId.+B$$$&@%#*+=~^7B[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]r}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFG\ .[]R}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()C4/1234567J90$+.[]{}()|\=1234567890$+.[]{}()|\/1234n6-890$+.[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@O$%^&*()_++I%###@%*+=*^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%m&*()_++I%###@\*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()\ I8***%#*\~^<>[]{}(B|\/qweXtyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8*i*%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwerYyuiopas+I8***%#*=~^<>[]{}()|\/qwertyyiopas9I8***%#*=~^<>[]{}()|\/q~ertyuiopas+I8***%#*=~^<>[]{}()|\/qwert_uiopas+I8**\ I&@@B#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbSmlkjhef+I&@@@#*+~^<>[h{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/z~cvbnmlkjhgf+I&@@@#*+~^<>n]{}()|\/zx|vbzmlkjhgf+5&@@\ I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+j^<>[]{}()|\/QWER$YUIOPASDF.I%$$$*0=^P>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$Ay*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QzERTYUIOPASDF+I%$$\ I8&&&+~^<>[]{}()|\/ZXCVBNMLKJH9FD+I8&&&+X^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|U/ZXCABNMLKJHGFD+I&&&&+~^<>[]{}()|\/4XCVBNML+JHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<3[\ I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>O]{}()|\/!@#$%^&*()_+-=[J+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+IE###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8***^<>[]{}()|\/`~1234567890&=2];+I8***^<>[]{}()|\/`~123456g8k0-=[];+I8***^<>[]{}()|\/`~1234567890-=j];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\&`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8**\ I&@@@<>[]{}()|\/'".,<>?%abcdefghij+I&@@@<>[]{}()|\/'"b,<>?/ab_defghij+I&@@J<>[]o}()|\/'".,<>?/gbcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdeBghij+I&@@@<>[]{}Z)|\/'".,<>?/abcdefgWij+I&@@@<>[]{}()|\/'".i<>?/abcdefgh-j+I&@@@<>[]{}()|\/'".,&>?/abcdefghij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEdGHIJKLMNOPQRSTE.<%$$[]{}()|~/ABCDENGHIJKLMCOPQRSG+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%E$[]{/()|\/ABCDEFGHIJKLMNOPQRST+.<%$\ .+B**{}()|\/12345.7890!@#$%^&*()_+.+BB*{9()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!1#$%H&*()h+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@@$%^&*()_+.+9**{}()|\\ .?%##()|\Aqwertyuiopasdfghjklzxcv+.?%##(E|\/qwertyuioMasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcr+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopDsdfghjklzxcv+.?9##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@s@@@.&@@@@@@@@@@@@@@@@@M@@@@@@@@@@@@@#@@@@.&@@t@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@u@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@|@@@@@\ @@@@@@@@@@0@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@a@@@@@@@@@@@@/@@@@@.@@@1@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@.@@@@@@@@@@@@@@s@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#7+=-~G<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{8()|\/a0cdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/a=cdefghijkl\ .+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+2$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~W<>[]{}()|\/A\CDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+%$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B8$$&@%#*+#~^<>[]f}()|\/ABCDEFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.~]{}()|\/1234567890$+.[]{}()|\N1234567890$+.[]d}()|\/1234567890$+.[]{}()|\/123456789\ I%###@%*+=~^<>[]{}()|\/!@#$Y^&*()_++I%###@%*+=~^<>[]{W()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!,#$%^&*()_++I%###@%*+=~^<>[]{}()|i/!g#$%^/*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8R**%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8**j%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*1~^m>[]{}()|\/qwertyuiopas+I8***%#Y=~^<>[]{}()|\/qwerty1iopas+I8***%#*=~^<>X]{}()|\/qwertyuio9as+I8**\ I&@1@#*+D^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbMmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+o&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ I%$$$*+=^<>[]{}()|\/QWERTjUIOPASDF+I%$$$*+=^<>[]I}()|\/QWERTYUIOPASDF@I%$$$*+=e<>[]{}()|\/QWERTYUIOPASDF+U%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$/*+=^[]{}r)|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDFMI%$$\ I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHG7D+I8&&&n~^<>[]{}()|\/ZXCVBNM&p8HGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()q\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()_+-5[]+I%3##=^[]{}()|\/!@#$%^&*()_+-=[]+I%###=^e>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$l^&*()_+-=[]+I%###=^E>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%##\ I8**d^<>[]{}()|\/`~1233567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8*x*^<>[]{}()|\/`~1234567890W=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~123456S890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1t*4567890-=S];+I8**\ I&@@@<>[]{}()|4/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abedefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRxT+.<%$$[]{}()|\/ABCDEFGHNJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}J)|\/ABCDEFCHIJKgMNaPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKL4NOPQRST+.<%$\ .+B**{}()|\/123456f890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*%)_+.+B**{}()|\/12345%7890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/b234567890!@#$%I&*()_+.+B**{}()|\\ .?%##()|\/qwertyuiopawdfghJklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/owertyuiopasdfghjklzxcv+.?%##()|\/qwertyufopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##(y|\/qwertyuiopasdfgHjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@-@@@@@@@@@@@9@@@@+@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@s@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@F@.@@@@@@@@@@@~@@@@@@@@@@@@@@@@@@@@a@,@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@L@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@A@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|~/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()o\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijkl\ .+B&$$&@%#*+=~^&>[]{}()|\/ABCDEAGHI+.+B$$$&@%#*+=~^<>s]{}()|\/A5CDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&Y%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[${}()|W/ABCDEFG\ .[]{}A)|\/1234567890$+w[]{}()|\/123456789Y$+.[]{}()|\/1234567890$6.[]{}()|\/G234567890$+.[]{}()|\/123456789o$+.[]{}()5\/1234567890$+.[]{}()|\/12345~~89\ I%###L%n+=~^<>[]{9()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}+K|\/!@#$x^&*()X++I%###@%*+=~^<>/]{}()|\/!@#$%^&*()_+DI%###@%*+=~^<>[]{}()|\&!@#$%^&*(pH++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~E<>[]{}()Q\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>I]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyui_p%s+I8***%#*=~^<>[]{}(\|\/qwertyuiopas+I8***%#*=~^f>K]{}()|\/qwertyuiopas+I8***%#*=~E<>[]{}()|\/qwertyuiopas+I8**\ I&@@@#*+~^<>u]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcv;nmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@\ 7%$$$*+=^<>[]{}()|\/QWERTYUIOPmRDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOP_SzF+Z%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]P}()|\/8WERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPAS+F+I%$$\ I8&&&+~C<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|e/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^Q>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[\ I%##+=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[-{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%^&*8)_+-=[]+I%##\ I8***^<>[]{}()n\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8**T^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{_()|\/`~1234567890-=[];+I8***^<>[]{}()|\1`~1234567890-=[];+v8***^<>[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[];+I8**\ I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'o.,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/aqcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".z<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}(z|\/ABCDEFGHIJKLMNOP:RST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}(S|\/ABCDEFGHIJKLMNOPQPST+.<%y$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$\ .+B**{}()|\/1234567890!@#$V^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+P+B**x}()|\/1234567890!@#B%^&*()_+.+B**{}()|\/12K4567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/123456789|!@#$%^&*()_-.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\\ .?%##()|\/qwertyuiopasdCghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyucopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##(C|\/qwertyuiopaidfghjklzxcv+.?%##()|\/qwurtyuiopasdfghjklzfcv+.?%##()|\Jq\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@G@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@8@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.9@@@@@@@@@@@@@@@@@@@@@@@r@@@@@@@@@@@@.&@@@@@@@@@I@@@@@@@@@@@@@@@@@@@Q@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@%@@@@@@@@@@@@@@@@@@@@@#@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@#@@@@@@@@@@@@@@@g@@@@@.@@@@@@@@@@@@~@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@V@@@@@@@@@@@.@@@@@@@@@@@@@@@4@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+w?%#*+=-~^<>[]{}L)|\/abcdefghijklmn+.?%#*+=-~^<>Z]{}()F\/abcdefghijklmn+.?k#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghiXklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}(0|\/abcdefghijkl\ .+B$$$&@%#*+=~^\>[]{}()|\/AUCDEWGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%@*+=B^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=~|<>[]{}()|\/ABCiEFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.o]{}()|\/1234567890$+.+C***+%#+=~^T>[]{}()|\/1234567890$+.[]{}()|\/1234567890$+.[]{}()|\/12W4567890$+.[]{}()|\/123456789\ I%###@%*+=~^<>P]{}()|\/!@#$%^&*()_++I%###@%*+=~^<#[]{}()|\/!@#x%^&*()_++I%###@%*+=~V<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}N)|\/!@#$%^&b()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^[]{}()|\/!@#$%^&*()\ I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyu.opas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I8***%#*=~^<>[]{}()|\/qwertyuiopas+I+***%#*=~^<>[]{}()|\/qwertKuiopas+I8**\ IM@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^l>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkRhgf+I&@@@#*+~^<>[]{}()|\/zxcvbnml|jhof+I&@@@#*+~^<>[]{}()|\/zxcvbnmlkjhgf+I&@@@#*+~^<>[]{2()|\/zxcvbn/lkjhgf+I&@@@#*+~^<>[]{}()|\/zxcibnmlkjhgf+I&@P\ I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$O+=^P>[]{}(E|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYU$OPASDF+I%$$$*+=^#>[]K}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTYUIOPASDF+I%$$$*+=^<>[]{}()|\/QWERTjUIOPpSDFtI%$#\ I8&&&+~^<>Z]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^Pd{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[v{}()|\/ZXCVBNMLKJHGFD+I8&&&+~^<>[]{}()|\mZ@CVBNMLKJHGFD+I8&&&+~^<>[]{}()|\/ZXCVBNMLKJHGFDdI/&&&+~^<>[\ I%###=^<>[]{}()|\/!@#$%^&*()Y+-=[]+I%###=^<>[]{}()|\/!@#$%^&*()_+-=[]+I%###=^<>[]{}()|\/!@#$%3&*()_+u=[]+I%###=^<>[]{}()|\/!@y$%^&*()_+-=[]+I%###=^<>X]{}()|\/!@#$%^_*()_+w=[]+I%###=^<>[]{}(,|\/!@#$%^&*()_+-=[]+I%###3^<>[]{}()|\/!@#$%^&*()_+\=[]+I%##\ I8***^<>[]{}()|\/`~1234567890-=[];+I8***^[]{}()|\/`~1234567890-=[];+I8***^<>[]{}()|\/`~1234567890-=[8;+I8***^2>[]{}()|\/`~1@34567890-=[];+I8***^<>[]{}()|\/`O1234567860-=[];+I8***^<>[]{}(T|\/`~1234567890-=z];+I8**\ I-@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@H@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghgj+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@@<>[]{}()2\/'".,f>?/abcdefghij+I&@@@<>[]{}()|\/'".,<>?/abcdefghij+I&@@\ .<%$$[]{}()|\zABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\7ABVDEFGHIJKLMNOPQRST+.<%v$[]{}()|\/ABCD@FGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCDEFGHIJKLMNOPQRST+.<%$$[]{}()|\/ABCD7FGHIJKLjNOPQWST+.<%$$[]{}()|\/ABCDEF+HIJKLMNOPQRSj+.<%$\ .+B**{}()|\/1234567890!@#$%^&*()_+.+B/*{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^I*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|\/123456789l!@#$%^&*()_+.+B**{}()|\/1234567890!@#$%^&*()_+.+B**{}()|w\ .?%##()|\Pqwertyuiopasdfghjklzxcv+.?%##()|\/qwertyu/opasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.W%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\\qwertyuiopasdfghjklzxcv+.?%##()|N/qwertyuiopasdfghjklzxcv+.?%##()|\/qwertyuiopasdfghjklzxcv+.?%##()|\/q\ &@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@S@@@.p@@@@@@@@@@@@@6@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@A@@@@@@@@@@@@@@@.&@@@@@@@@@@@@@@@@@@@@\ @@@@@@@@@@@@@@@@@@@s@@@@@@@@@@@@@@@@@Y@@@@@@@@q@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@F@@@@@@@E@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@X@m@@@Z@@L@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@.@@@@@@@@@@@@@@@@@@@@@\ .?%#*+=-~^<>[]{}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()j\/abcdefghijklmn+.?%#*+=$~^<>i]0}()|\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abcdefghijklmn6.?%#*+=-~^<>[]{}()2\/abcdefghijklmn+.?%#*+=-~^<>[]{}()|\/abc@efghijklmn+.?%#*+=-~^<>[]1}()|\/abcdefghijkl\ |+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+.+B$$$&@%#*+=i^<>[]{}3)|\/ABCDEFGHI+.+c$g$&@%#*+=~^<>[]{}()w\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDEFGHI+Q9B$$$&@%#*+=~^w>[]{}@)|\/ABCDEFGHI+.+B$$$&@%#*+=~^<>[]{}()|\/ABCDE*GHI+.+B$$$&@%#*+=~^<>[]{}()|\/P_CDzFG\ .[]{}()|\/1234567890$+.[]{}()|\/1234567890$X.[]{3()|\/1234567890$+.[]{}()|\/123~567890$+.[]{}()|\/1234567890$+.[]{}()|\+!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<+[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++\%###@%*+=~^<>[]{}Z)|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()_++I%###@%*+=~^<>[]{}()|\/!@#$%^&*()\ ```\ \ \ \ ## Join the community\ \ Connect with GPU kernel developers\ \ [JOIN DISCORD](https://discord.gg/sSJqgNnq6X)\ \ ## Now Available\ \ Download RightNow Editor\ \ Download our editor and start optimizing your CUDA kernels today.\ \ ## Windows\ \ Download\ \ Windows x64\ \ For Windows 11/10 x64\ \ Windows ARM\ \ For Windows 11/10 ARM\ \ Requires NVIDIA CUDA Toolkit [Download](https://developer.nvidia.com/cuda-downloads)\ \ ## Mac\ \ Download\ \ Mac Apple Silicon\ \ For M1, M2, M3, M4 Macs\ \ Mac Intel\ \ For Intel-based Macs\ \ Works on remote GPUs for free or our GPU emulator [Upgrade](/content/pricing/index.html)\ \ ## Linux\ \ Download\ \ Linux x64\ \ AppImage • Debian, Ubuntu, Fedora\ \ Linux ARM\ \ AppImage • ARM64 distributions\ \ Requires NVIDIA CUDA Toolkit [Download](https://developer.nvidia.com/cuda-downloads)