diff --git a/S1/uucoco_#36/HardELiSH_cuda.py b/S1/uucoco_#36/HardELiSH_cuda.py
new file mode 100644
index 00000000..faa2a229
--- /dev/null
+++ b/S1/uucoco_#36/HardELiSH_cuda.py
@@ -0,0 +1,77 @@
+import torch
+import torch.nn as nn
+from torch.utils.cpp_extension import load_inline
+
+class ModelNew(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self._compile_cuda_kernel()
+
+    def _compile_cuda_kernel(self):
+        cpp_source = """
+        torch::Tensor hard_elish_cuda(torch::Tensor x);
+        """
+
+        cuda_source = """
+        #include <torch/extension.h>
+        #include <cuda_runtime.h>
+        #include <math.h>
+
+        __device__ __forceinline__ float gate_op(float x) {
+            // max(0, min(1, (x + 1) / 2))
+            return fmaxf(0.0f, fminf(1.0f, x * 0.5f + 0.5f));
+        }
+
+        __device__ __forceinline__ float hard_elish_op(float x) {
+            float g = gate_op(x);
+            if (x >= 0.0f) {
+                return x * g;
+            } else {
+                // (exp(x) - 1) * g
+                return (expf(x) - 1.0f) * g;
+            }
+        }
+
+        __global__ void hard_elish_kernel(
+            const float* __restrict__ x,
+            float* __restrict__ output,
+            const int n_elements) 
+        {
+            const int block_start = blockIdx.x * blockDim.x;
+            const int block_end = min(block_start + blockDim.x, n_elements);
+
+            for (int i = block_start + threadIdx.x; i < block_end; i += blockDim.x) {
+                output[i] = hard_elish_op(x[i]);
+            }
+        }
+
+        torch::Tensor hard_elish_cuda(torch::Tensor x) {
+            auto x_c = x.contiguous();
+            const int n_elements = x_c.numel();
+            auto output = torch::empty_like(x_c);
+
+            const int threads = 256;
+            const int max_blocks = 65535;
+            const int blocks = std::min((n_elements + threads - 1) / threads, max_blocks);
+
+            hard_elish_kernel<<<blocks, threads>>>(
+                x_c.data_ptr<float>(),
+                output.data_ptr<float>(),
+                n_elements
+            );
+
+            return output;
+        }
+        """
+
+        self.op = load_inline(
+            name="hard_elish_v1",
+            cpp_sources=cpp_source,
+            cuda_sources=cuda_source,
+            functions=["hard_elish_cuda"],
+            extra_cuda_cflags=["-O3"],
+            verbose=False
+        )
+
+    def forward(self, x):
+        return self.op.hard_elish_cuda(x)
diff --git a/S1/uucoco_#36/HardELiSH_torch.py b/S1/uucoco_#36/HardELiSH_torch.py
new file mode 100644
index 00000000..79c44e02
--- /dev/null
+++ b/S1/uucoco_#36/HardELiSH_torch.py
@@ -0,0 +1,40 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Model(nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        gate = torch.clamp(x / 2.0 + 0.5, min=0.0, max=1.0)
+
+        positive_mask = x >= 0
+
+        output = torch.empty_like(x)
+
+        # x >= 0: x * gate
+        output_pos = x.masked_select(positive_mask) * gate.masked_select(positive_mask)
+        output.masked_scatter_(positive_mask, output_pos)
+
+        # x < 0: (exp(x) - 1) * gate
+        x_neg = x.masked_select(~positive_mask)
+        gate_neg = gate.masked_select(~positive_mask)
+        output_neg = (torch.exp(x_neg) - 1.0) * gate_neg
+        output.masked_scatter_(~positive_mask, output_neg)
+
+        return output
+
+
+batch_size = 512
+feature_dim = 1024
+
+
+def get_inputs():
+    x = torch.randn(batch_size, feature_dim, dtype=torch.float32)
+    return [x]
+
+
+def get_init_inputs():
+    return []
\ No newline at end of file
diff --git a/S1/uucoco_#36/prompt.txt b/S1/uucoco_#36/prompt.txt
new file mode 100644
index 00000000..c8027a62
--- /dev/null
+++ b/S1/uucoco_#36/prompt.txt
@@ -0,0 +1,57 @@
+You write custom CUDA kernels to replace the pytorch operators in the given GeGLU architecture to get speedups.
+
+You have complete freedom to choose the set of operators you want to replace. You may make the decision to replace some operators with custom CUDA kernels and leave others unchanged. You may replace multiple operators with custom implementations, consider operator fusion opportunities (combining multiple operators into a single kernel, for example, combining chunk+gelu+elementwise_mul), or algorithmic changes (such as optimized memory access patterns). You are only limited by your imagination.
+
+This CUDA kernel implements a custom Hard ELiSH activation function with the following optimizations:
+Tiled Kernel Design: Uses a block-based tiling approach where each thread processes elements within its assigned block, improving memory locality and cache efficiency compared to naive grid-stride loops.
+Branch Prediction Optimization: Implements conditional execution (x ≥ 0 vs x < 0) at the thread level, allowing warp-level parallelism to maintain efficiency despite the branch divergence.
+Fast Math Operations: Uses CUDA's fmaxf, fminf, and expfintrinsics for optimized mathematical computations without sacrificing readability.
+Memory Access Optimization: Employs __restrict__qualifiers and contiguous memory tensors to enable better compiler optimizations and reduce memory bank conflicts.
+Compiler Optimizations: Enabled with -O3flag for aggressive performance optimization of the generated machine code.
+Occupancy Optimization: Configures 256 threads per block and dynamically calculates grid size (up to 65535 blocks) to maximize GPU occupancy and resource utilization.
+Inlined Device Functions: Both the core activation function (hard_elish_op) and helper gate function (gate_op) are marked with __forceinline__to eliminate function call overhead within the kernel.
+Numerical Stability: Carefully implements the Hard ELiSH function with proper handling of the exponential term for negative inputs to maintain numerical precision.
+
+
+
+Here's an example to show you the syntax of inline embedding custom CUDA operators in torch: The example given architecture is:
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Model(nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        gate = torch.clamp(x / 2.0 + 0.5, min=0.0, max=1.0)
+
+        positive_mask = x >= 0
+
+        output = torch.empty_like(x)
+
+        # x >= 0: x * gate
+        output_pos = x.masked_select(positive_mask) * gate.masked_select(positive_mask)
+        output.masked_scatter_(positive_mask, output_pos)
+
+        # x < 0: (exp(x) - 1) * gate
+        x_neg = x.masked_select(~positive_mask)
+        gate_neg = gate.masked_select(~positive_mask)
+        output_neg = (torch.exp(x_neg) - 1.0) * gate_neg
+        output.masked_scatter_(~positive_mask, output_neg)
+
+        return output
+
+
+batch_size = 512
+feature_dim = 1024
+
+
+def get_inputs():
+    x = torch.randn(batch_size, feature_dim, dtype=torch.float32)
+    return [x]
+
+
+def get_init_inputs():
+    return []
\ No newline at end of file
diff --git a/S1/uucoco_#36/run_code.py b/S1/uucoco_#36/run_code.py
new file mode 100644
index 00000000..19d12213
--- /dev/null
+++ b/S1/uucoco_#36/run_code.py
@@ -0,0 +1,77 @@
+###########################################################
+# 性能和精度验证程序
+###########################################################
+import torch
+import torch.nn as nn
+import time
+from HardELiSH_torch import Model, get_inputs, get_init_inputs
+from HardELiSH_cuda  import ModelNew
+
+
+def run_benchmark():
+    # 检查 CUDA 是否可用
+    if not torch.cuda.is_available():
+        print("CUDA 不可用，请确保您有可用的 NVIDIA GPU 并已正确安装 PyTorch CUDA 版本。")
+        return
+    else:
+        device = torch.device("cuda")
+
+    # 初始化模型
+    init_inputs = get_init_inputs()
+    init_inputs = [
+        x.cuda(device=device) if isinstance(x, torch.Tensor) else x for x in init_inputs
+    ]
+    inputs = get_inputs()
+    inputs = [
+        x.cuda(device=device) if isinstance(x, torch.Tensor) else x for x in inputs
+    ]
+
+    torch_model = Model(*init_inputs).cuda()
+    cuda_model = ModelNew(*init_inputs).cuda()
+
+    torch_model.eval()
+    cuda_model.eval()
+
+    print("-------------------- 精度对齐验证 --------------------")
+    with torch.no_grad():
+        output_torch = torch_model(*inputs)
+        output_cuda = cuda_model(*inputs)
+
+    precision_flag = torch.allclose(output_torch, output_cuda, rtol=1e-03)
+    if precision_flag:
+        print("✅ 精度对齐：两个模型的输出结果非常接近。")
+    else:
+        print("❌ 精度不一致！")
+
+    print("\n-------------------- 性能加速比测试 --------------------")
+    num_iterations = 100
+
+    # PyTorch 模型计时
+    torch.cuda.synchronize()
+    start_time = time.time()
+    for _ in range(num_iterations):
+        _ = torch_model(*inputs)
+    torch.cuda.synchronize()
+    torch_time = (time.time() - start_time) / num_iterations
+
+    # 自定义 CUDA 内核计时
+    torch.cuda.synchronize()
+    start_time = time.time()
+    for _ in range(num_iterations):
+        _ = cuda_model(*inputs)
+    torch.cuda.synchronize()
+    cuda_time = (time.time() - start_time) / num_iterations
+
+    print(f"PyTorch torch.relu 平均执行时间: {torch_time:.6f} 秒")
+    print(f"自定义 CUDA 内核 平均执行时间: {cuda_time:.6f} 秒")
+    speedup = 0
+    if cuda_time > 0:
+        speedup = torch_time / cuda_time
+        print(f"加速比 (Speedup): {speedup:.2f}x")
+    else:
+        print("CUDA 内核执行时间为0，无法计算加速比。")
+    return precision_flag, speedup
+
+
+if __name__ == "__main__":
+    precision_flag, speedup = run_benchmark()
\ No newline at end of file