diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_infer.py b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_infer.py
new file mode 100644
index 0000000000000000000000000000000000000000..4b07552bad5c574b98c69c21c8ae1df6226676af
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_infer.py
@@ -0,0 +1,245 @@
+# Copyright 2023 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse 
+import os
+import random
+import time
+
+from PIL import Image
+from ais_bench.infer.interface import InferSession
+import cv2
+import einops
+import numpy as np
+from pytorch_lightning import seed_everything
+import torch
+
+from annotator.util import resize_image, HWC3
+from annotator.canny import CannyDetector
+from cldm.model import create_model, load_state_dict
+from cldm.ddim_hacked import DDIMSampler
+import config
+
+
+
+def parse_arguments():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--model",
+        type=str,
+        default="./models/control_sd15_canny.pth",
+        help="Path or name of the pre-trained model."
+    )
+    parser.add_argument(
+        "--image",
+        type=str,
+        default="./test_imgs/dog.png",
+        help="Path or name of the image."
+    )
+    parser.add_argument(
+        "--prompt",
+        type=str,
+        default="Prompt",
+        help="label=Prompt"
+    )
+    parser.add_argument(
+        "--a_prompt",
+        type=str,
+        default="best quality, extremely detailed",
+        help="added prompt"
+    )
+    parser.add_argument(
+        "--n_prompt",
+        type=str,
+        default="longbody, lowres, bad anatomy, bad hands, missing fingers, "
+                "extra digit, fewer digits, cropped, worst quality, low quality",
+        help="negative prompt"
+    )
+    parser.add_argument(
+        "--num_samples",
+        type=int,
+        default=1,
+        help="image_num"
+    )
+    parser.add_argument(
+        "--image_resolution",
+        type=int,
+        default=512,
+        help="image resolution"
+    )
+    parser.add_argument(
+        "--guess_mode",
+        type=bool,
+        default=False,
+        help="guess mode"
+    )
+    parser.add_argument(
+        "--strength",
+        type=float,
+        default=1.0,
+        help="control strength"
+    )
+    parser.add_argument(
+        "--scale",
+        type=float,
+        default=9.0,
+        help="guidance scale"
+    ) 
+    parser.add_argument(
+        "--seed",
+        type=int,
+        default=200,
+        help="seed"
+    )
+    parser.add_argument(
+        "--eta",
+        type=float,
+        default=0.0,
+        help="eta"
+    )
+    parser.add_argument(
+        "--low_threshold",
+        type=int,
+        default=100,
+        help="canny low threshold"
+    )
+    parser.add_argument(
+        "--high_threshold",
+        type=int,
+        default=200,
+        help="canny high threshold"
+    )
+    parser.add_argument(
+        "--control_model_dir",
+        type=str,
+        default="./models",
+        help="Base path of om models "
+    )
+    parser.add_argument(
+        "--sd_model_dir",
+        type=str,
+        default="./models",
+        help="Base path of om models."
+    )
+    parser.add_argument(
+        "--save_dir",
+        type=str,
+        default="./results",
+        help="Path to save result images."
+    )       
+    parser.add_argument(
+        "--ddim_steps",
+        type=int,
+        default=20,
+        help="Number of inference steps."
+    )
+    parser.add_argument(
+        "--device",
+        type=int,
+        default=0,
+        help="NPU device id."
+    )
+
+    return parser.parse_args()
+
+def process(model, ddim_sampler, sd_session, control_session, input_image, 
+            prompt, a_prompt, n_prompt, num_samples, image_resolution, 
+            ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, 
+            high_threshold,
+            ):
+    with torch.no_grad():
+        img = resize_image(HWC3(input_image), image_resolution)
+        H, W, C = img.shape
+
+        apply_canny = CannyDetector()
+        detected_map = apply_canny(img, low_threshold, high_threshold)
+        detected_map= HWC3(detected_map)
+
+        control = torch.from_numpy(detected_map.copy()).float().cpu() / 255.0
+        control = torch.stack([control for _ in range(1)],dim=0)
+        control = einops.rearrange(control, 'b h w c -> b c h w').clone()
+
+
+        if seed == -1:
+            seed = random.randint(0, 65535)
+        seed_everything(seed)
+
+        if config.save_memory:
+            model.low_vram_shift(is_diffusing=False)
+
+        cond = {"c_concat": [control], "c_crossattn": 
+                [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}
+        un_cond = {"c_concat": None if guess_mode else [control], 
+                   "c_crossattn": [model.get_learned_conditioning([n_prompt] * num_samples)]}
+        shape = (4, H // 8, W // 8)
+
+        if config.save_memory:
+            model.low_vram_shift(is_diffusing=True)
+
+        model.control_scales = [
+        	strength * (0.825 ** float(12 - i)) for i in range(13)] 
+        	if guess_mode else ([strength] * 13
+        )
+
+        samples, intermediates = ddim_sampler.sample(
+        	ddim_steps, num_samples,                  
+            shape, sd_session, control_session,
+            cond, verbose=False, eta=eta,
+            unconditional_guidance_scale=scale,
+            unconditional_conditioning=un_cond
+        )
+
+        if config.save_memory:
+            model.low_vram_shift(is_diffusing=False)
+
+        x_samples = model.decode_first_stage(samples)
+        x_samples = (
+        	einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + \
+        	127.5).cpu().numpy().clip(0, 255).astype(np.uint8)
+
+        results = [x_samples[i] for i in range(num_samples)]
+
+    return [255 - detected_map] + results
+
+def main():
+    args = parse_arguments()
+
+    model = create_model("./models/cldm_v15.yaml").cpu()
+    model.load_state_dict(load_state_dict(args.model, location="cpu"))
+    model = model.cpu()
+
+    ddim_sampler = DDIMSampler(model)
+
+    sd_om = args.sd_model_dir
+    control_om = args.control_model_dir
+
+    sd_session = InferSession(args.device, sd_om)
+    control_session = InferSession(args.device, control_om)
+
+    input_image = cv2.imread(args.image)
+    output = process(model, ddim_sampler, sd_session, control_session, input_image, 
+                     args.prompt, args.a_prompt, args.n_prompt, args.num_samples,
+                     args.image_resolution, args.ddim_steps, args.guess_mode,
+                     args.strength, args.scale, args.seed, args.eta, args.low_threshold,
+                     args.high_threshold)
+
+    if not os.path.exists(args.save_dir):
+        os.makedirs(args.save_dir, mode=0o744)
+    img0 = Image.fromarray(output[0])
+    img1 = Image.fromarray(output[1])
+    img0.save(os.path.join(args.save_dir, "cannyimg.png"))
+    img1.save(os.path.join(args.save_dir, "diffusionimg.png"))
+
+if __name__=="__main__":
+    main()
\ No newline at end of file
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_pth2onnx.py b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_pth2onnx.py
new file mode 100644
index 0000000000000000000000000000000000000000..3ffa007f481309ec051938524c714a0327c19f98
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/ControlNet_pth2onnx.py
@@ -0,0 +1,96 @@
+# Copyright 2023 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+import os
+
+import torch 
+from cldm.model import create_model, load_state_dict
+
+def parse_arguments():
+    parser = argparse.ArgumentParser()
+    parser.add_argument( 
+        "--model",
+        type=str,
+        default="./models/control_sd15_canny.pth",
+        help="Path or name of the pre-trained model.",
+    )
+    parser.add_argument(
+        "--control_path",
+        type=str,
+        default="./control.onnx",
+        help="path or name of the control.",
+    )
+    parser.add_argument(
+        "--sd_path",
+        type=str,
+        default="./sd.onnx",
+        help="Path or name of the sd.",
+    )
+
+    return parser.parse_args()
+
+def export_control(model, control_path):
+    model = model.control_model.eval()
+    dummy_input =(
+              torch.randn(1, 4, 64, 72),
+              torch.randn(1, 3, 512, 576),
+              torch.tensor([1]),
+              torch.randn(1, 77, 768)
+              )
+
+    torch.onnx.export(model, dummy_input, control_path, 
+                      input_names = ["text", "hint", "t","cond_text"],
+                      output_names = ["text_outs"], verbose=False, export_params=True,
+                      opset_version=13)
+
+def export_sd(model, sd_path):
+    model = model.model.diffusion_model.eval()
+    dummy_input = (
+               torch.randn(1, 4, 64, 72),
+               torch.tensor([1]),
+               torch.randn(1, 77, 768),
+               [torch.randn([1, 320, 64, 72]), torch.randn([1, 320, 64, 72]), 
+               torch.randn([1, 320, 64, 72]), torch.randn([1, 320, 32, 36]), 
+               torch.randn([1, 640, 32, 36]), torch.randn([1, 640, 32, 36]),
+               torch.randn([1, 640, 16, 18]), torch.randn([1, 1280, 16, 18]), 
+               torch.randn([1, 1280, 16, 18]), torch.randn([1, 1280, 8, 9]),
+               torch.randn([1, 1280, 8, 9]), torch.randn([1, 1280, 8, 9]),
+               torch.randn([1, 1280, 8, 9])]
+               )
+
+    torch.onnx.export(model, dummy_input, sd_path, 
+                      input_names = ["text", "t", "cond_text",
+                                     "input1", "input2", "input3", "input4",
+                                     "input5", "input6", "input7", "input8",
+                                     "input9", "input10", "input11", "input12", 
+                                     "input13"],
+                      output_names = ["text_outs"],
+                      verbose=False, export_params=True, opset_version=13)
+
+def main():
+    args = parse_arguments()
+    model = create_model("./models/cldm_v15.yaml" ).cpu()
+    model.load_state_dict(load_state_dict(args.model, location='cpu'))
+    if not os.path.exists(args.control_path):
+        os.makedirs(args.control_path, mode=0o744)
+    if not os.path.exists(args.sd_path):
+        os.makedirs(args.sd_path, mode=0o744)
+    export_control(model, os.path.join(args.control_path, "control.onnx"))
+    print("control model done")
+    export_sd(model, os.path.join(args.sd_path, "sd.onnx"))
+    print("sd model done")
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/README.md b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..194116606e21d90399f445f05b49700b2c2bf9f9
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/README.md
@@ -0,0 +1,254 @@
+# ControlNet模型-推理指导  
+
+
+- [概述](#ZH-CN_TOPIC_0000001172161501)
+   
+   - [输入输出数据](#section540883920406)
+
+- [推理环境准备](#ZH-CN_TOPIC_0000001126281702)
+
+- [快速上手](#ZH-CN_TOPIC_0000001126281700)
+
+  - [获取源码](#section4622531142816)
+  - [准备数据集](#section183221994411)
+  - [模型推理](#section741711594517)
+
+- [模型推理性能&精度](#ZH-CN_TOPIC_0000001172201573)
+
+
+# 概述<a name="ZH-CN_TOPIC_0000001172161501"></a>
+
+   ControlNet是一种神经网络结构，通过添加额外的条件来控制扩散模型，它将神经网络块的权重复制到“锁定”副本和“可训练”副本中。“可训练”的人会了解您的病情。“锁定”的模型会保留您的模型。因此，使用图像对的小数据集进行训练不会破坏生产就绪的扩散模型。“零卷积”是 1×1 卷积，权重和偏差都初始化为零。在训练之前，所有零卷积都输出零,ControlNet 不会造成任何失真。这允许在小规模甚至个人设备上进行培训。这也有利于合并/替换/偏移模型/权重/块/层。
+
+- 参考实现：
+  ```
+  url=https://github.com/lllyasviel/ControlNet 
+  branch=main
+  commit_id=ed85cd1e25a5ed592f7d8178495b4483de0331bf
+  ```
+
+## 输入输出数据<a name="section540883920406"></a>
+
+- 输入数据
+
+  | 输入数据  | 大小      | 数据类型                | 数据排布格式 |
+  | -------- | -------- | ------------------------- | ------------ |
+  | text    |  1 x 4 x 64 x 72 | FLOAT32 |  NCHW|
+  | hint    |  1 x 3 x 512 x 576 | FLOAT32 | NCHW|
+  | t       |  1                | INT64 | ND|
+  | cond_text| 1 x 77 x 768     | FLOAT32| ND|
+
+
+- 输出数据
+
+  | 输出数据 | 大小     | 数据类型 | 数据排布格式 |
+  | -------- | -------- | -------- | ------------ |
+  | text_outs | 1 x 4 x 64 x 72 | FLOAT32  | NCHW           |
+
+# 推理环境准备<a name="ZH-CN_TOPIC_0000001126281702"></a>
+
+- 该模型需要以下插件与驱动
+
+  **表 1**  版本配套表
+  | 配套                                                         | 版本    | 环境准备指导                                                 |
+  | ------------------------------------------------------------ | ------- | ------------------------------------------------------------ |
+  | 固件与驱动                                                   | 23.0.rc3  | [Pytorch框架推理环境准备](https://www.hiascend.com/document/detail/zh/ModelZoo/pytorchframework/pies) |
+  | CANN                                                         | 7.0.0 | -                                                            |
+  | Python                                                       | 3.8.5   | -                                                            |                                                           |
+
+
+# 快速上手<a name="ZH-CN_TOPIC_0000001126281700"></a>
+
+## 获取源码<a name="section4622531142816"></a>
+
+1. 获取源码
+   ```
+   git clone https://github.com/lllyasviel/ControlNet
+   mv ControlNet_pth2onnx.py ControlNet_infer.py pipeline.py ControlNet/
+   ```
+
+2. 安装依赖。
+   ```
+   pip3 install -r requirements.txt
+   ```
+
+3. 代码修改
+
+   执行命令：
+   
+   ```
+   patch -p1 < differences.patch
+   ```
+
+4. 安装昇腾统一推理工具（AIT）
+
+   请访问[AIT代码仓](https://gitee.com/ascend/ait/tree/master/ait#ait)，根据readme文档进行工具安装。
+
+   安装AIT时，可只安装需要的组件：benchmark，其他组件为可选安装。
+   
+## 准备数据集<a name="section183221994411"></a>
+
+1. 获取原始数据集。
+
+   本模型输入原始图片和文本生成图片，无需数据集。
+   
+## 模型推理<a name="section741711594517"></a>
+
+1. 模型转换。
+   使用PyTorch将模型权重文件.pth转换为.onnx文件，再使用ATC工具将.onnx文件转为离线推理模型文件.om文件。
+
+   1. 获取权重
+
+     
+      ```
+      训练权重链接为："https://huggingface.co/lllyasviel/ControlNet/blob/main/models/control_sd15_canny.pth"。
+      下载后放入`ControlNet/models`工作目录下.
+      ```
+
+   2. 导出ONNX模型
+
+      
+      ```
+      可提前下载openclip权重放入'ControlNet/openai/clip-vit-large-patch14'，以避免执行后面步骤时可能会出现下载失败。
+      # 需要使用 git-lfs (https://git-lfs.com)
+      git lfs install
+      cd ControlNet
+      git clone https://huggingface.co/openai/clip-vit-large-patch14
+      ```
+
+      
+
+      执行命令：
+
+      ```
+      python ControlNet_pth2onnx.py --model ./models/control_sd15_canny.pth --control_path onnx/control/  --sd_path onnx/sd/
+      ```
+
+      参数说明：
+      - --model：本地模型目录的路径
+      - --control_path: control部分ONNX模型输出目录
+      - --sd_path：sd部分onnx模型输出目录
+    
+
+   
+   3. 使用ATC工具将ONNX模型转OM模型。
+
+      1. 配置环境变量。
+
+         ```
+         source /usr/local/Ascend/ascend-toolkit/set_env.sh
+         ```
+
+         > **说明：** 
+         >该脚本中环境变量仅供参考，请以实际安装环境配置环境变量。详细介绍请参见《[CANN 开发辅助工具指南 \(推理\)](https://support.huawei.com/enterprise/zh/ascend-computing/cann-pid-251168373?category=developer-documents&subcategory=auxiliary-development-tools)》。
+
+      2. 执行命令查看芯片名称（$\{chip\_name\}）。
+
+         ```
+         npu-smi info
+         #该设备芯片名为Ascend310P3 （自行替换）
+         回显如下：
+         +-------------------+-----------------+------------------------------------------------------+
+         | NPU     Name      | Health          | Power(W)     Temp(C)           Hugepages-Usage(page) |
+         | Chip    Device    | Bus-Id          | AICore(%)    Memory-Usage(MB)                        |
+         +===================+=================+======================================================+
+         | 0       310P3     | OK              | 15.8         42                0    / 0              |
+         | 0       0         | 0000:82:00.0    | 0            1074 / 21534                            |
+         +===================+=================+======================================================+
+         | 1       310P3     | OK              | 15.4         43                0    / 0              |
+         | 0       1         | 0000:89:00.0    | 0            1070 / 21534                            |
+         +===================+=================+======================================================+
+         ```
+
+      3. 执行ATC命令。
+
+         ```
+         # control
+         atc --framework=5 \
+             --model=./onnx/control/control.onnx \
+             --output=./om/control/control \
+             --input_format=ND \
+             --log=error \
+             --soc_version=Ascend${chip_name}
+         
+         # sd
+         atc --framework=5 \
+             --model=./onnx/sd/sd.onnx \
+             --output=./om/sd/sd \
+             --input_format=ND \
+             --log=error \
+             --soc_version=Ascend${chip_name}
+
+         ```
+      
+      参数说明：
+      - --model：为ONNX模型文件。
+      - --output：输出的OM模型。
+      - --framework：5代表ONNX模型。
+      - --log：日志级别。
+      - --soc_version：处理器型号。
+
+2. 开始推理验证。
+
+   1. 执行推理脚本。
+      ```
+    
+      python3 ControlNet_infer.py \
+              --model ./models/control_sd15_canny.pth \
+              --image test_imgs/dog.png
+              --prompt "cute dog" \
+              --device 0 \
+              --control_model_dir om/control/control.om \
+              --sd_model_dir om/sd/sd.om \
+              --save_dir ./results \
+              --ddim_steps 20 
+   
+      ```
+
+      参数说明：
+      - --model：本地模型目录的路径。
+      - --prompt：文本信息。
+      - --save_dir：生成图片的存放目录。
+      - --ddim_steps：生成图片次数。
+      - --image: 输入图片。
+      - --control_model_dir: control的om位置。
+      - --sd_model_dir: sd的om位置。
+      - --device：推理设备ID。
+      
+      执行完成后在`./results`目录下生成推理图片。推理一张图片会输出一张图片边缘的图片，和一张跟据输入图片和文本重新生成的图片。
+
+3. 推理结果图片展示。
+
+   在线模型推理结果：
+
+   ![](./test_results/dog0.PNG)
+
+   离线模型推理结果：
+
+   ![](./test_results/dog1.PNG)
+   
+
+4. 性能验证。
+
+      可使用ais_bench推理工具的纯推理模式验证om模型的性能，参考命令如下：
+
+        ```
+
+        python -m ais_bench --model=${om_model_path} --loop=10 --batchsize=${batch_size} --device 0,1
+        ```
+
+      - 参数说明：
+        - --model：模型路径
+        - --batchsize：每批次样本数量
+        - --device:Duo卡测试时设两张芯片
+
+   
+# 模型推理性能&精度<a name="ZH-CN_TOPIC_0000001172201573"></a>
+
+调用ACL接口推理计算，性能参考下列数据。
+
+
+| 硬件形态 | 性能 | 模型|
+| :------: | :--------: | :--------: |
+| Duo并行  | 33.6fps  | control |  
+| Duo并行 | 13.6fps | sd |
\ No newline at end of file
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/differences.patch b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/differences.patch
new file mode 100644
index 0000000000000000000000000000000000000000..5a1b189ba4a3f75f4f6431bc99a5a62fd0526e01
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/differences.patch
@@ -0,0 +1,125 @@
+diff -Naru a/ControlNet/cldm/ddim_hacked.py b/ControlNet/cldm/ddim_hacked.py
+--- a/ControlNet/cldm/ddim_hacked.py	2023-11-17 23:27:48.688000000 +0800
++++ b/ControlNet/cldm/ddim_hacked.py	2023-11-17 23:15:45.548000000 +0800
+@@ -16,8 +16,8 @@
+ 
+     def register_buffer(self, name, attr):
+         if type(attr) == torch.Tensor:
+-            if attr.device != torch.device("cuda"):
+-                attr = attr.to(torch.device("cuda"))
++            if attr.device != torch.device("cpu"):
++                attr = attr.to(torch.device("cpu"))
+         setattr(self, name, attr)
+ 
+     def make_schedule(self, ddim_num_steps, ddim_discretize="uniform", ddim_eta=0., verbose=True):
+@@ -56,6 +56,8 @@
+                S,
+                batch_size,
+                shape,
++               sd_session,
++               control_session,
+                conditioning=None,
+                callback=None,
+                normals_sequence=None,
+@@ -101,6 +103,8 @@
+         print(f'Data shape for DDIM sampling is {size}, eta {eta}')
+ 
+         samples, intermediates = self.ddim_sampling(conditioning, size,
++                                                    sd_session,
++                                                    control_session,
+                                                     callback=callback,
+                                                     img_callback=img_callback,
+                                                     quantize_denoised=quantize_x0,
+@@ -120,7 +124,7 @@
+         return samples, intermediates
+ 
+     @torch.no_grad()
+-    def ddim_sampling(self, cond, shape,
++    def ddim_sampling(self, cond, shape, sd_session, control_session,
+                       x_T=None, ddim_use_original_steps=False,
+                       callback=None, timesteps=None, quantize_denoised=False,
+                       mask=None, x0=None, img_callback=None, log_every_t=100,
+@@ -160,7 +164,7 @@
+                 assert len(ucg_schedule) == len(time_range)
+                 unconditional_guidance_scale = ucg_schedule[i]
+ 
+-            outs = self.p_sample_ddim(img, cond, ts, index=index, use_original_steps=ddim_use_original_steps,
++            outs = self.p_sample_ddim(img, cond, ts, sd_session, control_session, index=index, use_original_steps=ddim_use_original_steps,
+                                       quantize_denoised=quantize_denoised, temperature=temperature,
+                                       noise_dropout=noise_dropout, score_corrector=score_corrector,
+                                       corrector_kwargs=corrector_kwargs,
+@@ -178,7 +182,7 @@
+         return img, intermediates
+ 
+     @torch.no_grad()
+-    def p_sample_ddim(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
++    def p_sample_ddim(self, x, c, t, sd_session, control_session, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
+                       temperature=1., noise_dropout=0., score_corrector=None, corrector_kwargs=None,
+                       unconditional_guidance_scale=1., unconditional_conditioning=None,
+                       dynamic_threshold=None):
+@@ -187,8 +191,8 @@
+         if unconditional_conditioning is None or unconditional_guidance_scale == 1.:
+             model_output = self.model.apply_model(x, t, c)
+         else:
+-            model_t = self.model.apply_model(x, t, c)
+-            model_uncond = self.model.apply_model(x, t, unconditional_conditioning)
++            model_t = self.model.apply_model(x, t, c, sd_session, control_session)
++            model_uncond = self.model.apply_model(x, t, unconditional_conditioning, sd_session, control_session)
+             model_output = model_uncond + unconditional_guidance_scale * (model_t - model_uncond)
+ 
+         if self.model.parameterization == "v":
+diff -Naru a/ControlNet/ldm/modules/attention.py b/ControlNet/ldm/modules/attention.py
+--- a/ControlNet/ldm/modules/attention.py	2023-11-17 23:27:49.192000000 +0800
++++ b/ControlNet/ldm/modules/attention.py	2023-11-17 23:17:33.896000000 +0800
+@@ -174,7 +174,7 @@
+         if _ATTN_PRECISION =="fp32":
+             with torch.autocast(enabled=False, device_type = 'cuda'):
+                 q, k = q.float(), k.float()
+-                sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
++                sim = einsum('bid,bjd->bij', q, k) * self.scale
+         else:
+             sim = einsum('b i d, b j d -> b i j', q, k) * self.scale
+         
+@@ -189,7 +189,7 @@
+         # attention, what we cannot get enough of
+         sim = sim.softmax(dim=-1)
+ 
+-        out = einsum('b i j, b j d -> b i d', sim, v)
++        out = einsum('bij,bjd->bid', sim, v)
+         out = rearrange(out, '(b h) n d -> b n (h d)', h=h)
+         return self.to_out(out)
+ 
+diff -Naru a/ControlNet/ldm/modules/diffusionmodules/util.py b/ControlNet/ldm/modules/diffusionmodules/util.py
+--- a/ControlNet/ldm/modules/diffusionmodules/util.py	2023-11-17 23:27:49.192000000 +0800
++++ b/ControlNet/ldm/modules/diffusionmodules/util.py	2023-11-17 23:23:05.404000000 +0800
+@@ -109,7 +109,7 @@
+                    explicitly take as arguments.
+     :param flag: if False, disable gradient checkpointing.
+     """
+-    if flag:
++    if flag and not torch.onnx.is_in_onnx_export():
+         args = tuple(inputs) + tuple(params)
+         return CheckpointFunction.apply(func, len(inputs), *args)
+     else:
+diff -Naru a/ControlNet/ldm/modules/encoders/modules.py b/ControlNet/ldm/modules/encoders/modules.py
+--- a/ControlNet/ldm/modules/encoders/modules.py	2023-11-17 23:27:49.192000000 +0800
++++ b/ControlNet/ldm/modules/encoders/modules.py	2023-11-17 23:20:02.000000000 +0800
+@@ -92,7 +92,7 @@
+         "pooled",
+         "hidden"
+     ]
+-    def __init__(self, version="openai/clip-vit-large-patch14", device="cuda", max_length=77,
++    def __init__(self, version="openai/clip-vit-large-patch14", device="cpu", max_length=77,
+                  freeze=True, layer="last", layer_idx=None):  # clip-vit-base-patch32
+         super().__init__()
+         assert layer in self.LAYERS
+diff -Naru a/ControlNet/models/cldm_v15.yaml b/ControlNet/models/cldm_v15.yaml
+--- a/ControlNet/models/cldm_v15.yaml	2023-11-17 23:27:49.196000000 +0800
++++ b/ControlNet/models/cldm_v15.yaml	2023-11-17 23:18:39.812000000 +0800
+@@ -1,5 +1,5 @@
+ model:
+-  target: cldm.cldm.ControlLDM
++  target: pipeline.AscendControlNet
+   params:
+     linear_start: 0.00085
+     linear_end: 0.0120
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/pipeline.py b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/pipeline.py
new file mode 100644
index 0000000000000000000000000000000000000000..388df82f2f1c50399a50122aed814c4a075a69b6
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/pipeline.py
@@ -0,0 +1,46 @@
+# Copyright 2023 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import torch
+from cldm.cldm import ControlLDM
+
+class AscendControlNet(ControlLDM):
+    def apply_model(self, x_noisy, t, cond, sd_session, control_session):
+        assert isinstance(cond, dict)
+        cond_txt = torch.cat(cond["c_crossattn"], 1)
+        mode = "static"
+
+        control = control_session.infer(
+                        [
+                            x_noisy.numpy(),
+                            torch.cat(cond["c_concat"], 1).numpy(),                            
+                            t.numpy(),
+                            cond_txt.numpy()
+                        ], mode
+                    )
+
+        control = [c * scale for c, scale in zip(control, self.control_scales)]
+
+        eps = torch.from_numpy(
+                sd_session.infer(
+                [
+                    x_noisy.numpy(),
+                    t.numpy(),
+                    cond_txt .numpy()
+                ]               
+                + control, mode
+                )[0]
+        )
+
+        return eps
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/requirements.txt b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b53c8a9faa68136e7f0f23d36fa8ab8985ac5118
--- /dev/null
+++ b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/requirements.txt
@@ -0,0 +1,24 @@
+torch==1.12.1
+torchvision==0.13.1
+numpy==1.23.1
+albumentations==1.3.0
+opencv-python-headless==4.8.1.78
+imageio==2.9.0
+imageio-ffmpeg==0.4.2
+pytorch-lightning==1.5.0
+omegaconf==2.1.1
+test-tube>=0.7.5
+streamlit==1.12.1
+einops==0.3.0
+transformers==4.19.2
+webdataset==0.2.5
+kornia==0.6
+open_clip_torch==2.0.2
+invisible-watermark>=0.1.5
+streamlit-drawable-canvas==0.8.0
+torchmetrics==0.6.0 
+timm==0.6.12
+addict==2.4.0
+yapf==0.32.0
+prettytable==3.6.0
+safetensors==0.2.7
\ No newline at end of file
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog0.PNG b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog0.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..3cfe8346ddf339ef4a1cfb276a752f3d5a05e1b1
Binary files /dev/null and b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog0.PNG differ
diff --git a/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog1.PNG b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog1.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..3cfe8346ddf339ef4a1cfb276a752f3d5a05e1b1
Binary files /dev/null and b/ACL_PyTorch/built-in/foundation_models/ControlNet_for_PyTorch/test_results/dog1.PNG differ