diff --git a/docs/vllm_mindspore/docs/source_en/conf.py b/docs/vllm_mindspore/docs/source_en/conf.py
index 964df2041b7bcb9d840850d8c3f915588fe5158b..f82c63fa1cd975e9dce5f67e2cb99c30ca1c30c0 100644
--- a/docs/vllm_mindspore/docs/source_en/conf.py
+++ b/docs/vllm_mindspore/docs/source_en/conf.py
@@ -37,7 +37,7 @@ copyright = 'MindSpore'
 author = 'vLLM-MindSpore Plugin'
 
 # The full version, including alpha/beta/rc tags
-release = '0.5.0'
+release = '0.5.1'
 
 # -- General configuration ---------------------------------------------------
 
diff --git a/docs/vllm_mindspore/docs/source_en/developer_guide/contributing.md b/docs/vllm_mindspore/docs/source_en/developer_guide/contributing.md
index 52dda0064de6a79e50a96c208c2c11a7305933c3..1a4deadc42328d5238843426a40c423fdad175b9 100644
--- a/docs/vllm_mindspore/docs/source_en/developer_guide/contributing.md
+++ b/docs/vllm_mindspore/docs/source_en/developer_guide/contributing.md
@@ -18,7 +18,7 @@ To support a new model for vLLM-MindSpore Plugin code repository, please note th
 - **Follow file format and location specifications.** Model code files should be placed under the `vllm_mindspore/model_executor` directory, organized in corresponding subfolders by model type.
 - **Implement models using MindSpore interfaces with jit static graph support.** Model definitions in vLLM-MindSpore Plugin must be implemented using MindSpore interfaces. Since MindSpore's static graph mode offers performance advantages, models should support execution via @jit static graphs. For reference, see the [Qwen2.5](https://gitee.com/mindspore/vllm-mindspore/blob/r0.5.0/vllm_mindspore/model_executor/models/qwen2.py) implementation.
 - **Register new models in vLLM-MindSpore Plugin.** After implementing the model structure, register it in vLLM-MindSpore Plugin by adding it to `_NATIVE_MODELS` in `vllm_mindspore/model_executor/models/registry.py`.
-- **Write unit tests.** New models must include corresponding unit tests. Refer to the [Qwen2.5 testcases](https://gitee.com/mindspore/vllm-mindspore/blob/r0.5.0/tests/st/python/cases_parallel/vllm_qwen_7b.py) for examples.
+- **Write unit tests.** New models must include corresponding unit tests. Refer to the [Qwen2.5 testcases](https://gitee.com/mindspore/vllm-mindspore/blob/master/tests/st/models/qwen2_5/test_vllm_qwen_7b_v1.py) for examples.
 
 ## Contribution Process
 
diff --git a/docs/vllm_mindspore/docs/source_en/faqs/faqs.md b/docs/vllm_mindspore/docs/source_en/faqs/faqs.md
index 24343451c885e943e07d5d7d50f05bd6c01a1a89..fdb8fa789889587f8f1520f9bea8d4e6bccc8e49 100644
--- a/docs/vllm_mindspore/docs/source_en/faqs/faqs.md
+++ b/docs/vllm_mindspore/docs/source_en/faqs/faqs.md
@@ -11,7 +11,7 @@
     Execute the following command to install vLLM-MindSpore Plugin:
 
     ```bash
-    git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+    git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
     cd vllm-mindspore
     bash install_depend_pkgs.sh
     pip install .
@@ -29,7 +29,7 @@
     2. Please check if the `pip` version is greater than or equal to 25.3. If so, please use the following command to compile and install vLLM-MindSpore Plugin:
 
         ```bash
-        git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+        git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
         cd vllm-mindspore
         bash install_depend_pkgs.sh
         pip install --no-build-isolation .
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/installation/installation.md b/docs/vllm_mindspore/docs/source_en/getting_started/installation/installation.md
index c185b14f7abd6360aa94410c48ca83eb6bbf1a12..8774ce195d2bcae00ca00c5f9a5a7b42dd535b59 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/installation/installation.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/installation/installation.md
@@ -10,18 +10,24 @@ This document will introduce the [Version Matching](#version-compatibility) of v
 ## Version Compatibility
 
 - OS: Linux-aarch64
-- Python：3.10 / 3.11 / 3.12
+- Python：3.9 / 3.10 / 3.11
 - Depent Software version compatibility
 
    | Software | Version And Links |
    | -----    | -----   |
    | CANN  |   [8.3.RC1](https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/83RC1/index/index.html)      |
-   | MindSpore |  [2.7.1.post1](https://repo.mindspore.cn/mindspore/mindspore/version/202512/20251225/r2.7.1_20251225204822_6eefcb41b2c30e42eb049908d5efea40c3425abc_newest/)    |
+   | MindSpore |  [2.7.1.post1](https://repo.mindspore.cn/mindspore/mindspore/version/202512/20251229/r2.7.1_20251229100307_d3099307fd78df39099f0e7801a6d2aa3121bf2c_newest/)    |
    | MSAdapter| [0.0.5](https://repo.mindspore.cn/mindspore/msadapter/version/202510/20251011/r0.3.0_20251011095813_951a8218d4c29785e48f304e720212b57056573e_newest/) |
-   | MindSpore Transformers | [1.7.0](https://repo.mindspore.cn/mindspore/mindformers/version/202510/20251030/r1.7.0_20251030031507_8ccc49b3f6645d3d1abfab80b4c78f3cafe5c84e_newest/)  |
-   | vLLM     | [0.11.0](https://repo.mindspore.cn/mirrors/vllm/version/202511/20251113/v0.11.0/) |
-   | ms_custom_ops | [0.1.0](https://repo.mindspore.cn/mindspore/ms_custom_ops/version/202512/20251210/r0.1.0_20251210100500_15d84de93ca616c1a9880275c15213a34f2c0a39_newest/) |
-   | MindSpore ONE  | [0.5.0](https://repo.mindspore.cn/mindspore-lab/mindone/version/202512/20251223/master_20251223115818_36c25653b3ebaff55eb40ca10bc742d603b21f0a_newest/) |
+   | MindSpore Transformers | [1.7.0](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1/MindFormers/any/mindformers-1.7.0-py3-none-any.whl)  |
+   | vLLM     | [0.11.0](https://repo.mindspore.cn/mirrors/vllm/version/202511/20251113/v0.11.0/any/vllm-0.11.0%2Bempty-py3-none-any.whl) |
+   | ms_custom_ops | [0.1.0](https://repo.mindspore.cn/mindspore/ms_custom_ops/version/202512/20251229/r0.1.0_20251229145109_5bd05d81cf153ce96c2244fbf5d0c03b9fba1ffe_newest/) |
+   | MindSpore ONE  | [0.5.0](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/MindOne/any/mindone-0.5.0-py3-none-any.whl) |
+
+- Source code and download link of vLLM-MindSpore Plugin
+
+   | Source Code Link | Package Link |
+   | -----    | -----   |
+   | [v0.5.1](https://gitee.com/mindspore/vllm-mindspore/tree/v0.5.1/) | [Python3.9](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp39-cp39-linux_aarch64.whl)，[Python3.10](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp310-cp310-linux_aarch64.whl)，[Python3.11](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp311-cp311-linux_aarch64.whl) |   
 
 ## Docker Installation
 
@@ -32,7 +38,7 @@ We recommend using Docker for quick deployment of the vLLM-MindSpore Plugin envi
 User can execute the following commands to clone the vLLM-MindSpore Plugin code repository:
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```  
 
 To build the image according to your npu type, follow these steps:
@@ -136,7 +142,7 @@ vLLM-MindSpore Plugin can be installed in the following two ways. **vLLM-MindSpo
     To install vLLM-MindSpore Plugin, user needs to pull the vLLM-MindSpore Plugin source code and then runs the following command to install the dependencies:
 
     ```bash
-    git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+    git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
     cd vllm-mindspore
     bash install_depend_pkgs.sh
     ```
@@ -206,7 +212,7 @@ vLLM-MindSpore Plugin can be installed in the following two ways. **vLLM-MindSpo
         User could also use source code to install vLLM-MindSpore Plugin.
 
         ```bash
-        git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+        git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
         cd vllm-mindspore
         pip install .
         ```
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/quick_start/quick_start.md b/docs/vllm_mindspore/docs/source_en/getting_started/quick_start/quick_start.md
index 6b0b5f5c6ee3a2e6a01904c4fcf1edb5b24bdd82..48422e84d6db815c9b860fe9b6b50734fc5a6b63 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/quick_start/quick_start.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/quick_start/quick_start.md
@@ -13,7 +13,7 @@ In this section, we recommend using docker to deploy the vLLM-MindSpore Plugin e
 User can execute the following commands to clone the vLLM-MindSpore Plugin code repository:
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```  
 
 To build the image according to your npu type, follow these steps:
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
index a22a7dec49922f30d32d0b550ccadaa713bc729f..c0ae90dc0ad7b643484ed5cf5b99509c9713faeb 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
@@ -17,7 +17,7 @@ In this section, we recommend using Docker creation to quickly deploy the vLLM-M
 Users can execute the following commands to pull the vLLM-MindSpore plugin code repository and build the image:
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 bash build_image.sh
 ```
 
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
index f7847be4f55650a944e6d9c639d97568e72c4f1d..738b35abd539f1a336b21aef61a851b0a622628c 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
@@ -13,7 +13,7 @@ In this section, we recommend using Docker for quick deployment of the vLLM-Mind
 User can execute the following commands to clone the vLLM-MindSpore Plugin code repository:
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```  
 
 To build the image according to your npu type, follow these steps:
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
index c9dea90b9875341b9356d1bd9d44304e821ccbd0..7c326924fcb68343d3f71b6eed2dbb3f644cff61 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
@@ -13,7 +13,7 @@ In this section, we recommend using Docker for quick deployment of the vLLM-Mind
 User can execute the following commands to clone the vLLM-MindSpore Plugin code repository:
 
 ```bash  
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```  
 
 To build the image according to your npu type, follow these steps:
diff --git a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
index 7549a292f64e7dba66f83d7e7484d412ebfe4ef7..8911a06199a71e94c46d08e28c33d82ab24ebd23 100644
--- a/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
+++ b/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
@@ -1,8 +1,8 @@
-# Multi-modal Single-Card Inference (Qwen2.5-7B)
+# Multi-modal Single-Card Inference (Qwen3-VL-8B-Instruct)
 
 [![View Source On Gitee](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.7.1.post1/resource/_static/logo_source_en.svg)](https://gitee.com/mindspore/docs/blob/r2.7.1.post1/docs/vllm_mindspore/docs/source_en/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md)  
 
-This document introduces single NPU inference process by vLLM-MindSpore Plugin. Taking the [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) model as an example, user can configure the environment through the [Docker Installation](#docker-installation) or the [Installation Guide](../../installation/installation.md#installation-guide), and [downloading model weights](#downloading-model-weights). After [setting environment variables](#setting-environment-variables), user can perform [offline inference](#offline-inference) and [online inference](#online-inference) to experience single NPU inference abilities.
+This document introduces single NPU multimodal inference process by vLLM-MindSpore Plugin. Taking the [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) model as an example, user can configure the environment through the [Docker Installation](#docker-installation) or the [Installation Guide](../../installation/installation.md#installation-guide), and [downloading model weights](#downloading-model-weights). After [setting environment variables](#setting-environment-variables), user can perform [offline inference](#offline-inference) and [online inference](#online-inference) to experience single NPU inference abilities.
 
 ## Docker Installation
 
@@ -13,7 +13,7 @@ In this section, we recommend using Docker for quick deployment of the vLLM-Mind
 User can execute the following commands to clone the vLLM-MindSpore Plugin code repository:
 
 ```bash  
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```  
 
 To build the image according to your npu type, follow these steps:
@@ -92,22 +92,7 @@ docker exec -it $DOCKER_NAME bash
 
 ## Downloading Model Weights
 
-User can download the model using either [Python Tool](#downloading-with-python-tool) or [git-lfs Tool](#downloading-with-git-lfs-tool).  
-
-### Downloading with Python Tool
-
-Execute the following Python script to download the [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) weights and files from [Hugging Face](https://huggingface.co/):  
-
-```python  
-from huggingface_hub import snapshot_download  
-snapshot_download(  
-    repo_id="Qwen/Qwen3-VL-8B-Instruct",  
-    local_dir="/path/to/save/Qwen3-VL-8B-Instruct",  
-    local_dir_use_symlinks=False  
-)  
-```  
-
-`local_dir` is the user-specified model save path. Ensure sufficient disk space is available.  
+User can download the model using either Web or [git-lfs Tool](#downloading-with-git-lfs-tool).  
 
 ### Downloading with git-lfs Tool
 
@@ -137,7 +122,7 @@ For [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct), th
 
 ```bash  
 #set environment variables  
-export VLLM_MS_MODEL_BACKEND=MindFormers # use MindSpore TransFormers as model backend.
+export VLLM_MS_MODEL_BACKEND=Native # use Native Model as model backend.
 ```  
 
 Here is an explanation of these variables:  
@@ -162,14 +147,11 @@ from vllm import LLM, SamplingParams
 # Sample prompts.
 PROMPT_TEMPLATE = (
     "<|im_start|>system\nYou are a helpful assistant.<|im_end|>"
-    "\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>"
-    "Is there anyone in the picture?<|im_end|>\n"
+    "\n<|im_start|>user\nDescribe the content of the image"
+    "<|vision_start|><|image_pad|><|vision_end|><|im_end|>\n"
     "<|im_start|>assistant\n")
 
-# example:
-# https://tools.mindspore.cn/dataset/workspace/mindspore_dataset/images/houses_and_mountain.jpeg
-
-image_path = "/path/to/houses_and_mountain.jpeg"
+image_path = "/path/to/image.jpg"
 
 def pil_image() -> Image.Image:
     return Image.open(image_path)
@@ -184,13 +166,15 @@ inputs = [
 ]
 
 # Create a sampling params object.
-sampling_params = SamplingParams(temperature=0.0, top_p=0.95)
+sampling_params = SamplingParams(max_tokens=512)
 model_path = "/path/to/save/Qwen3-VL-8B-Instruct"
 # Create a LLM
-llm = LLM(model=model_path)
+llm = LLM(model=model_path，
+          max_model_len=32768,
+          gpu_memory_utilization=0.85)
 # Generate texts from the prompts. The output is a list of RequestOutput objects
 # that contain the prompt, generated text, and other information.
-outputs = llm.generate(prompts, sampling_params)
+outputs = llm.generate(inputs, sampling_params)
 # Print the outputs.
 for output in outputs:
     prompt = output.prompt
@@ -201,7 +185,7 @@ for output in outputs:
 If offline inference runs successfully, similar results will be obtained:
 
 ```text
-Generated text: This is a breathtaking landscape photograph capturing a serene, idyllic scene in a mountainous region. Here's a breakdown of what's in the image:\n\n*   **Foreground:** A vibrant green meadow dotted with wildflowers, including yellow dandelions and some
+"Generated text: This image captures a Pallas's cat (also known as the manul or Otocolobus manul) walking through a snowy landscape.\n\nKey features of the cat:\n* Distinctive Appearance: The cat has a stocky, robust build with a very thick, dense coat of fur that is a mix of gray, brown, and buff colors. This thick fur, dusted with snowflakes, is a key adaptation for its cold, high-altitude habitat.\n* Round Head and Face: It has a wide, rounded head and a remarkably flattened face with short, broad muzzle and large, prominent ears.\n* Gaze: The cat is looking down and to its left, seemingly focused on something in the snow.\n* Posture: It is walking or stalking with one front paw lifted, indicating movement.\n* Paws: Its paws are large and furry, which help it walk on snow.\n\nEnvironment:\n* The cat is on a blanket of fresh white snow.\n* The background is composed of the characteristic white, peeling bark of birch trees and a dark, possibly wrought iron, fence.\n* The lighting suggests an overcast day, which is common in winter mountain environments where this species lives.\n\nIn summary, the image provides a clear, naturalistic view of a Pallas's cat navigating its snowy, wooded habitat, showcasing its unique and endearing physical characteristics."
 ```
 
 ## Online Inference
@@ -213,7 +197,7 @@ vLLM-MindSpore Plugin supports online inference deployment with the OpenAI API p
 Start the vLLM service with the following command:
 
 ```bash
-nohup vllm-mindspore serve /path/to/save/Qwen3-VL-8B-Instruct &
+nohup vllm-mindspore serve /path/to/save/Qwen3-VL-8B-Instruct --max-model-len 32768 --gpu-memory-utilization 0.85 &
 ```
 
 User can also set the local model path as model tag. If the service starts successfully, similar output will be obtained:
@@ -251,17 +235,13 @@ def send_request():
     )
     print(response.json())
 
-# The image for encoding
-# example:
-# https://tools.mindspore.cn/dataset/workspace/mindspore_dataset/images/houses_and_mountain.jpeg
-
-image_path = "/path/to/houses_and_mountain.jpeg"
+image_path = "/path/to/image.jpg"
 
 base64_image = encode_image(image_path)
 
 # make a request
 payload = {
-    "model": "/home/ckpt/Qwen2.5-VL-7B-Instruct",
+    "model": "/home/ckpt/Qwen3-VL-8B-Instruct",
     "messages": [
         {
             "role": "user",
@@ -270,13 +250,19 @@ payload = {
                     "type": "image_url",
                     "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"}
                 },
-                {"type": "text", "text": "Is there anyone in the picture?"}
+                {"type": "text", "text": "Describe the content of the image"}
             ]
         }
     ],
-    "max_tokens": 100,
-    "temperature": 0.1,
+    "max_tokens": 512,
 }
+
+send_request()
 ```
 
 User needs to ensure that the `"model"` field matches the model tag in the service startup, and the request can successfully match the model.
+
+If online inference runs successfully, similar results will be obtained:
+```
+{'id': 'chatcmpl-0f0a85dcbe7343f89539200e1a201e04', 'object': 'chat.completion', 'created': 1769408795, 'model': '/home/ckpt/Qwen3-VL-8B-Instruct', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': 'This is a photograph of a Pallas's cat (also known as a "manul") walking through a snowy landscape.\n\nHere are the key details:\n\n* Subject: The central focus is a Pallas's cat, a small wild feline native to Central Asia. It is covered in thick, fluffy fur that is a mix of gray, brown, and tan, with distinct dark markings around its eyes and on its cheeks.\n* Action: The cat is captured mid-stride, walking forward through the snow. Its body is low to the ground, and its front left paw is lifted, indicating movement.\n* Environment: The setting is a winter scene. The ground is covered in white snow, and the background consists of the distinctive white, peeling bark of birch trees. There are also some dark, vertical elements in the background, possibly a fence or another structure.\n* Atmosphere: The image conveys a sense of quiet wilderness. The cat's thick fur is dusted with snow, suggesting it has been moving through the snow for some time. The overall mood is calm and natural.\n\nThe Pallas's cat's unique, somewhat "pug-faced" appearance, with its large, rounded ears and dense coat, is clearly visible, making it a striking and memorable subject.', 'refusal': None, 'annotations': None, 'audio': None, 'function_call': None, 'tool_calls': [], 'reasoning_content': None}, 'logprobs': None, 'finish_reason': 'stop', 'stop_reason': None, 'token_ids': None}], 'service_tier': None, 'system_fingerprint': None, 'usage': {'prompt_tokens': 646, 'total_tokens': 917, 'completion_tokens': 271, 'prompt_tokens_details': None}, 'prompt_logprobs': None, 'prompt_token_ids': None, 'kv_transfer_params': None}
+```
diff --git a/docs/vllm_mindspore/docs/source_en/index.rst b/docs/vllm_mindspore/docs/source_en/index.rst
index 351e36ab42063d494de64ef204cbed4eb264c9a5..de5f5edb335cbca1c09a9a19506a8a681777486c 100644
--- a/docs/vllm_mindspore/docs/source_en/index.rst
+++ b/docs/vllm_mindspore/docs/source_en/index.rst
@@ -39,8 +39,8 @@ Prerequisites
 - Operating System: openEuler or Ubuntu Linux
 - Software:
 
-  * Python >= 3.10, <= 3.12
-  * CANN >= 8.3.RC1
+  * Python >= 3.9, <= 3.11
+  * CANN == 8.3.RC1
   * MindSpore (matched with the vllm-mindspore version)
   * vLLM (matched with the vllm-mindspore version)
 
@@ -88,9 +88,9 @@ The following are the version branches:
    *  -  r0.4.0
       -  Maintained
       -  Compatible with vLLM v0.9.1, and CI commitment for MindSpore 2.7.1
-   *  -  r0.5.0
+   *  -  v0.5.1
       -  Maintained
-      -  Compatible with vLLM v0.11.0, and CI commitment for MindSpore 2.7.1.post1   
+      -  Compatible with vLLM v0.11.0, and CI commitment for MindSpore 2.7.1.post1
 
 SIG
 -----------------------------------------------------
diff --git a/docs/vllm_mindspore/docs/source_en/release_notes/release_notes.md b/docs/vllm_mindspore/docs/source_en/release_notes/release_notes.md
index 0e75a38251452cde7c18a25a8a669035327b7001..d95e6acbce46cf33252dead3b62293f768705940 100644
--- a/docs/vllm_mindspore/docs/source_en/release_notes/release_notes.md
+++ b/docs/vllm_mindspore/docs/source_en/release_notes/release_notes.md
@@ -2,9 +2,9 @@
 
 [![View Source On Gitee](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.7.1.post1/resource/_static/logo_source_en.svg)](https://gitee.com/mindspore/docs/blob/r2.7.1.post1/docs/vllm_mindspore/docs/source_en/release_notes/release_notes.md)
 
-## vLLM-MindSpore Plugin 0.5.0 Release Notes
+## vLLM-MindSpore Plugin 0.5.1 Release Notes
 
-The vLLM-MindSpore Plugin 0.5.0 version is compatible with vLLM 0.11.0. Below are the new features and models supported in this release.
+The vLLM-MindSpore Plugin 0.5.1 version is compatible with vLLM 0.11.0. Below are the new features and models supported in this release.
 
 ### New Features
 
@@ -18,6 +18,13 @@ The vLLM-MindSpore Plugin 0.5.0 version is compatible with vLLM 0.11.0. Below ar
 - GLM Model Series:
     - GLM-4.1V Native Model
 
+### Problem Fixes
+
+- Encoder Memory Leak Issue
+    - [Fix for vLLM](https://github.com/vllm-project/vllm/pull/31857)
+    - [Fix for vLLM-MindSpore Plugin](https://gitee.com/mindspore/vllm-mindspore/pulls/1447)
+
+
 ### Contributors
 
 Thanks to the following contributors for their efforts:
diff --git a/docs/vllm_mindspore/docs/source_en/user_guide/supported_models/models_list/models_list.md b/docs/vllm_mindspore/docs/source_en/user_guide/supported_models/models_list/models_list.md
index 7b657dab5d6a03c0882762f41c190ab344eacf65..ef548f0fc515a81e997847c4068a0518dc601576 100644
--- a/docs/vllm_mindspore/docs/source_en/user_guide/supported_models/models_list/models_list.md
+++ b/docs/vllm_mindspore/docs/source_en/user_guide/supported_models/models_list/models_list.md
@@ -10,7 +10,7 @@
 | DeepSeek-R1 W8A4 |   Supported | MindFormers | Atlas 800I A2 | [DeepSeek-R1-W8A4](https://modelers.cn/models/MindSpore-Lab/DeepSeek-R1-0528-gs-A8W4) |
 | Telechat2 | Supported | MindFormers | Atlas 800I A2 | [TeleChat2-7B-32K](https://www.modelscope.cn/models/TeleAI/TeleChat2-7B-32K), [TeleChat2-35B-32K](https://www.modelscope.cn/models/TeleAI/TeleChat2-35B-32K) |
 | GLM-4.5 | Supported | MindFormers | Atlas 800I A2 | [GLM-4.5](https://huggingface.co/zai-org/GLM-4.5), [GLM-4.5-Air](https://huggingface.co/zai-org/GLM-4.5-Air) |
-| GLM-4.1V-Thinking | Supported | MindFormers | Atlas 800I A2 | [GLM-4.1V-9B-Thinking](https://huggingface.co/zai-org/GLM-4.1V-9B-Thinking) |
+| GLM-4.1V-Thinking | Supported | Native | Atlas 800I A2 | [GLM-4.1V-9B-Thinking](https://huggingface.co/zai-org/GLM-4.1V-9B-Thinking) |
 | GLM-4 | Supported | MindFormers | Atlas 800I A2 | [GLM-4-9B-0414](https://huggingface.co/zai-org/GLM-4-9B-0414), [GLM-4-32B-0414](https://huggingface.co/zai-org/GLM-4-32B-0414) |
 | Qwen3-VL | Supported | Native  | Atlas 800I A2 | [Qwen3-VL-2B](https://huggingface.co/Qwen/Qwen3-VL-2B-Instruct), [Qwen3-VL-4B](https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct), [Qwen3-VL-8B](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct), [Qwen3-VL-30B-A3B](https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct), [Qwen3-VL-235B-A22B](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Instruct) |
 | Qwen3 | Supported | Native, MindFormers | Atlas 800I A2, Atlas 300I Duo | [Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B), [Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B), [Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B), [Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B), [Qwen3-14B](https://modelers.cn/models/MindSpore-Lab/Qwen3-14B), [Qwen3-32B](https://modelers.cn/models/MindSpore-Lab/Qwen3-32B) |
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/conf.py b/docs/vllm_mindspore/docs/source_zh_cn/conf.py
index ca970f28a83f04c1789035affe247a8f5b213f6d..a2e5503c37a6942cc8b8786b8443a3fe1d069b2c 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/conf.py
+++ b/docs/vllm_mindspore/docs/source_zh_cn/conf.py
@@ -28,7 +28,7 @@ copyright = 'MindSpore'
 author = 'vLLM-MindSpore Plugin'
 
 # The full version, including alpha/beta/rc tags
-release = '0.5.0'
+release = '0.5.1'
 
 # -- General configuration ---------------------------------------------------
 
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/developer_guide/contributing.md b/docs/vllm_mindspore/docs/source_zh_cn/developer_guide/contributing.md
index 2351177758e66ca6d76f4bd8f99e344dcaf769cb..7da84784bfff455557f3d495f7ecfb420ce4d180 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/developer_guide/contributing.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/developer_guide/contributing.md
@@ -19,7 +19,7 @@
 - **文件格式及位置要遵循规范。** 模型代码文件统一放置于`vllm_mindspore/model_executor`文件夹下，请根据不同模型将代码文件放置于对应的文件夹下。
 - **模型基于MindSpore接口实现，支持jit静态图方式执行。** vLLM-MindSpore插件中的模型定义需基于MindSpore接口实现。由于MindSpore静态图模式执行性能有优势，因此模型需支持@jit静态图方式执行。详细可参考[Qwen2.5](https://gitee.com/mindspore/vllm-mindspore/blob/r0.5.0/vllm_mindspore/model_executor/models/qwen2.py)模型定义实现。
 - **将新模型在vLLM-MindSpore插件代码中进行注册。** 模型结构定义实现后，需要将该模型注册到vLLM-MindSpore插件中，注册文件位于'vllm_mindspore/model_executor/models/registry.py'中，请将模型注册到`_NATIVE_MODELS`。
-- **编写单元测试。** 新增的模型需同步提交单元测试用例，用例编写请参考[Qwen2.5模型用例](https://gitee.com/mindspore/vllm-mindspore/blob/r0.5.0/tests/st/python/cases_parallel/vllm_qwen_7b.py)。
+- **编写单元测试。** 新增的模型需同步提交单元测试用例，用例编写请参考[Qwen2.5模型用例](https://gitee.com/mindspore/vllm-mindspore/blob/master/tests/st/models/qwen2_5/test_vllm_qwen_7b_v1.py)。
 
 ## 贡献流程
 
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/faqs/faqs.md b/docs/vllm_mindspore/docs/source_zh_cn/faqs/faqs.md
index 71b34071d5a8e9c2056de2eaf9908ed1b6ae5a79..5df7f9c832ad91e7788b8446681fe0c4dfe305a3 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/faqs/faqs.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/faqs/faqs.md
@@ -11,7 +11,7 @@
     在执行以下命令安装vLLM-MindSpore Plugin
 
     ```bash
-    git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+    git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
     cd vllm-mindspore
     bash install_depend_pkgs.sh
     pip install .
@@ -29,7 +29,7 @@
     2. 请检查`pip`版本是否大于等于25.3。如果是，则使用以下命令编译并安装vLLM-MindSpore Plugin：
 
         ```bash
-        git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+        git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
         cd vllm-mindspore
         bash install_depend_pkgs.sh
         pip install --no-build-isolation .
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/installation/installation.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/installation/installation.md
index ab35b88911223fc1507f9edd1686437339f39c1c..3127218de36b242d17cf08136ec676684f6b9ae5 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/installation/installation.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/installation/installation.md
@@ -10,18 +10,24 @@
 ## 版本配套
 
 - OS：Linux-aarch64
-- Python：3.10 / 3.11 / 3.12
+- Python：3.9 / 3.10 / 3.11
 - 依赖软件版本配套
 
    | 软件 | 配套版本与下载链接 |
    | -----    | -----   |
    | CANN  |   [8.3.RC1](https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/83RC1/index/index.html)      |
-   | MindSpore |  [2.7.1.post1](https://repo.mindspore.cn/mindspore/mindspore/version/202512/20251225/r2.7.1_20251225204822_6eefcb41b2c30e42eb049908d5efea40c3425abc_newest/)    |
+   | MindSpore |  [2.7.1.post1](https://repo.mindspore.cn/mindspore/mindspore/version/202512/20251229/r2.7.1_20251229100307_d3099307fd78df39099f0e7801a6d2aa3121bf2c_newest/)    |
    | MSAdapter| [0.0.5](https://repo.mindspore.cn/mindspore/msadapter/version/202510/20251011/r0.3.0_20251011095813_951a8218d4c29785e48f304e720212b57056573e_newest/) |
-   | MindSpore Transformers | [1.7.0](https://repo.mindspore.cn/mindspore/mindformers/version/202510/20251030/r1.7.0_20251030031507_8ccc49b3f6645d3d1abfab80b4c78f3cafe5c84e_newest/)  |
-   | vLLM     | [0.11.0](https://repo.mindspore.cn/mirrors/vllm/version/202511/20251113/v0.11.0/) |
-   | ms_custom_ops | [0.1.0](https://repo.mindspore.cn/mindspore/ms_custom_ops/version/202512/20251210/r0.1.0_20251210100500_15d84de93ca616c1a9880275c15213a34f2c0a39_newest/) |
-   | MindSpore ONE  | [0.5.0](https://repo.mindspore.cn/mindspore-lab/mindone/version/202512/20251223/master_20251223115818_36c25653b3ebaff55eb40ca10bc742d603b21f0a_newest/) |
+   | MindSpore Transformers | [1.7.0](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1/MindFormers/any/mindformers-1.7.0-py3-none-any.whl)  |
+   | vLLM     | [0.11.0](https://repo.mindspore.cn/mirrors/vllm/version/202511/20251113/v0.11.0/any/vllm-0.11.0%2Bempty-py3-none-any.whl) |
+   | ms_custom_ops | [0.1.0](https://repo.mindspore.cn/mindspore/ms_custom_ops/version/202512/20251229/r0.1.0_20251229145109_5bd05d81cf153ce96c2244fbf5d0c03b9fba1ffe_newest/) |
+   | MindSpore ONE  | [0.5.0](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/MindOne/any/mindone-0.5.0-py3-none-any.whl) |
+
+- vLLM-MindSpore插件代码仓与下载链接
+
+   |代码仓链接 | 插件包下载链接 |
+   | -----    | -----   |
+   | [v0.5.1](https://gitee.com/mindspore/vllm-mindspore/tree/v0.5.1/) | [Python3.9](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp39-cp39-linux_aarch64.whl)，[Python3.10](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp310-cp310-linux_aarch64.whl)，[Python3.11](https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.1.post1/VllmMindSpore/ascend/aarch64/vllm_mindspore-0.5.1-cp311-cp311-linux_aarch64.whl) |
 
 ## docker安装
 
@@ -32,7 +38,7 @@
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```
 
 根据计算卡类型，构建镜像：
@@ -136,7 +142,7 @@ vLLM-MindSpore插件有以下两种安装方式。**vLLM-MindSpore插件快速
     采用快速安装脚本来安装vLLM-MindSpore插件，需要在拉取vLLM-MindSpore插件源码后，执行以下命令安装依赖包：
 
     ```bash
-    git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+    git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
     cd vllm-mindspore
     bash install_depend_pkgs.sh
     ```
@@ -206,7 +212,7 @@ vLLM-MindSpore插件有以下两种安装方式。**vLLM-MindSpore插件快速
         也可以进行如下的源码编译安装：
 
         ```bash
-        git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+        git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
         cd vllm-mindspore
         pip install .
         ```
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/quick_start/quick_start.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/quick_start/quick_start.md
index 1c4097ac335c08d2a69a050621ce073e5e4f2ab4..4363e674268f7d49398d2619ce38db8f725db055 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/quick_start/quick_start.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/quick_start/quick_start.md
@@ -13,7 +13,7 @@
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```
 
 根据计算卡类型，构建镜像：
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
index 441c7f00f93702ed4154ea74352ba154086e9329..0d843e63614759d6c876be313149372076e05695 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/deepseek_parallel/deepseek_r1_671b_w8a8_dp4_tp4_ep4.md
@@ -17,7 +17,7 @@ vLLM-MindSpore插件支持[张量并行（TP）](https://docs.vllm.ai/en/v0.11.0
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库，并构建镜像：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 bash build_image.sh
 ```
 
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
index c4b9cc0d80fdbe620b7ed8334874101ab2a6b3a4..5f86f1469581b8bbe4e581611694e6daab7d5dcc 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_32b_multiNPU/qwen2.5_32b_multiNPU.md
@@ -13,7 +13,7 @@
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```
 
 根据计算卡类型，构建镜像：
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
index aac0d4ef9212b1fd5f4c0a03850b47630be8eb9c..ce2d86e1e4d06fee5f925cb9a9c918ee81bf12de 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen2.5_7b_singleNPU/qwen2.5_7b_singleNPU.md
@@ -13,7 +13,7 @@
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```
 
 根据计算卡类型，构建镜像：
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
index 929357b206c3f54f4c63341f643d5a913c32b4ce..05785b03cd64316471b0118a5eda919c8c69f874 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md
@@ -2,7 +2,7 @@
 
 [![查看源文件](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source.svg)](https://gitee.com/mindspore/docs/blob/r2.7.1.post1/docs/vllm_mindspore/docs/source_zh_cn/getting_started/tutorials/qwen3_vl_8b_singleNPU/qwen3_vl_8b_singleNPU.md)
 
-本文档将介绍使用vLLM-MindSpore插件进行单卡推理的流程。以[Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)模型为例，用户可通过以下[docker安装](#docker安装)章节或[安装指南](../../installation/installation.md#安装指南)章节进行环境配置，并[下载模型权重](#下载模型权重)。在[设置环境变量](#设置环境变量)之后，可进行[离线推理](#离线推理)与[在线推理](#在线推理)，体验单卡推理功能。
+本文档将介绍使用vLLM-MindSpore插件进行多模态单卡推理的流程。以[Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)模型为例，用户可通过以下[docker安装](#docker安装)章节或[安装指南](../../installation/installation.md#安装指南)章节进行环境配置，并[下载模型权重](#下载模型权重)。在[设置环境变量](#设置环境变量)之后，可进行[离线推理](#离线推理)与[在线推理](#在线推理)，体验单卡推理功能。
 
 ## docker安装
 
@@ -13,7 +13,7 @@
 用户可执行以下命令，拉取vLLM-MindSpore插件代码仓库：
 
 ```bash
-git clone -b r0.5.0 https://gitee.com/mindspore/vllm-mindspore.git
+git clone -b v0.5.1 https://gitee.com/mindspore/vllm-mindspore.git
 ```
 
 根据计算卡类型，构建镜像：
@@ -92,22 +92,7 @@ docker exec -it $DOCKER_NAME bash
 
 ## 下载模型权重
 
-用户可采用[Python工具下载](#python工具下载)或[git-lfs工具下载](#git-lfs工具下载)两种方式，进行模型下载。
-
-### Python工具下载
-
-执行以下 Python 脚本，从[Hugging Face社区](https://huggingface.co/)下载[Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)权重及文件：
-
-```python
-from openmind_hub import snapshot_download
-snapshot_download(
-    repo_id="Qwen/Qwen3-VL-8B-Instruct",
-    local_dir="/path/to/save/Qwen3-VL-8B-Instruct",
-    local_dir_use_symlinks=False
-)
-```
-
-其中`local_dir`为模型保存路径，由用户指定，请确保该路径下有足够的硬盘空间。
+用户可采用Hugging Face网页下载或者(#git-lfs工具下载)两种方式，进行模型下载。
 
 ### git-lfs工具下载
 
@@ -162,14 +147,11 @@ from vllm import LLM, SamplingParams
 # Sample prompts.
 PROMPT_TEMPLATE = (
     "<|im_start|>system\nYou are a helpful assistant.<|im_end|>"
-    "\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>"
-    "Is there anyone in the picture?<|im_end|>\n"
+    "\n<|im_start|>user\nDescribe the content of the image"
+    "<|vision_start|><|image_pad|><|vision_end|><|im_end|>\n"
     "<|im_start|>assistant\n")
 
-# example:
-# https://tools.mindspore.cn/dataset/workspace/mindspore_dataset/images/houses_and_mountain.jpeg
-
-image_path = "/path/to/houses_and_mountain.jpeg"
+image_path = "/path/to/image.jpg"
 
 def pil_image() -> Image.Image:
     return Image.open(image_path)
@@ -184,13 +166,15 @@ inputs = [
 ]
 
 # Create a sampling params object.
-sampling_params = SamplingParams(temperature=0.0, top_p=0.95)
+sampling_params = SamplingParams(max_tokens=512)
 model_path = "/path/to/save/Qwen3-VL-8B-Instruct"
 # Create a LLM
-llm = LLM(model=model_path)
+llm = LLM(model=model_path，
+          max_model_len=32768,
+          gpu_memory_utilization=0.85)
 # Generate texts from the prompts. The output is a list of RequestOutput objects
 # that contain the prompt, generated text, and other information.
-outputs = llm.generate(prompts, sampling_params)
+outputs = llm.generate(inputs, sampling_params)
 # Print the outputs.
 for output in outputs:
     prompt = output.prompt
@@ -201,7 +185,7 @@ for output in outputs:
 若成功执行，则可以获得类似的执行结果：
 
 ```text
-Generated text: This is a breathtaking landscape photograph capturing a serene, idyllic scene in a mountainous region. Here's a breakdown of what's in the image:\n\n*   **Foreground:** A vibrant green meadow dotted with wildflowers, including yellow dandelions and some
+"Generated text: This image captures a Pallas's cat (also known as the manul or Otocolobus manul) walking through a snowy landscape.\n\nKey features of the cat:\n* Distinctive Appearance: The cat has a stocky, robust build with a very thick, dense coat of fur that is a mix of gray, brown, and buff colors. This thick fur, dusted with snowflakes, is a key adaptation for its cold, high-altitude habitat.\n* Round Head and Face: It has a wide, rounded head and a remarkably flattened face with short, broad muzzle and large, prominent ears.\n* Gaze: The cat is looking down and to its left, seemingly focused on something in the snow.\n* Posture: It is walking or stalking with one front paw lifted, indicating movement.\n* Paws: Its paws are large and furry, which help it walk on snow.\n\nEnvironment:\n* The cat is on a blanket of fresh white snow.\n* The background is composed of the characteristic white, peeling bark of birch trees and a dark, possibly wrought iron, fence.\n* The lighting suggests an overcast day, which is common in winter mountain environments where this species lives.\n\nIn summary, the image provides a clear, naturalistic view of a Pallas's cat navigating its snowy, wooded habitat, showcasing its unique and endearing physical characteristics."
 ```
 
 ## 在线推理
@@ -213,7 +197,7 @@ vLLM-MindSpore插件可使用OpenAI的API协议，部署在线推理。以下以
 使用如下命令启动vLLM服务：
 
 ```bash
-nohup vllm-mindspore serve /path/to/save/Qwen3-VL-8B-Instruct &
+nohup vllm-mindspore serve /path/to/save/Qwen3-VL-8B-Instruct --max-model-len 32768 --gpu-memory-utilization 0.85 &
 ```
 
 用户可以通过指定模型保存的本地路径作为模型标签。若服务成功启动，则可以获得类似的执行结果：
@@ -251,17 +235,13 @@ def send_request():
     )
     print(response.json())
 
-# 编码图片
-# example:
-# https://tools.mindspore.cn/dataset/workspace/mindspore_dataset/images/houses_and_mountain.jpeg
-
-image_path = "/path/to/houses_and_mountain.jpeg"
+image_path = "/path/to/image.jpg"
 
 base64_image = encode_image(image_path)
 
 # 构造请求
 payload = {
-    "model": "/home/ckpt/Qwen2.5-VL-7B-Instruct",
+    "model": "/home/ckpt/Qwen3-VL-8B-Instruct",
     "messages": [
         {
             "role": "user",
@@ -270,13 +250,19 @@ payload = {
                     "type": "image_url",
                     "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"}
                 },
-                {"type": "text", "text": "Is there anyone in the picture?"}
+                {"type": "text", "text": "Describe the content of the image"}
             ]
         }
     ],
-    "max_tokens": 100,
-    "temperature": 0.1,
+    "max_tokens": 512,
 }
+
+send_request()
 ```
 
 其中，用户需确认`"model"`字段与启动服务中的模型标签一致，请求才能成功匹配到模型。
+
+若成功执行，则可以获得类似的执行结果：
+```
+{'id': 'chatcmpl-0f0a85dcbe7343f89539200e1a201e04', 'object': 'chat.completion', 'created': 1769408795, 'model': '/home/ckpt/Qwen3-VL-8B-Instruct', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': 'This is a photograph of a Pallas's cat (also known as a "manul") walking through a snowy landscape.\n\nHere are the key details:\n\n* Subject: The central focus is a Pallas's cat, a small wild feline native to Central Asia. It is covered in thick, fluffy fur that is a mix of gray, brown, and tan, with distinct dark markings around its eyes and on its cheeks.\n* Action: The cat is captured mid-stride, walking forward through the snow. Its body is low to the ground, and its front left paw is lifted, indicating movement.\n* Environment: The setting is a winter scene. The ground is covered in white snow, and the background consists of the distinctive white, peeling bark of birch trees. There are also some dark, vertical elements in the background, possibly a fence or another structure.\n* Atmosphere: The image conveys a sense of quiet wilderness. The cat's thick fur is dusted with snow, suggesting it has been moving through the snow for some time. The overall mood is calm and natural.\n\nThe Pallas's cat's unique, somewhat "pug-faced" appearance, with its large, rounded ears and dense coat, is clearly visible, making it a striking and memorable subject.', 'refusal': None, 'annotations': None, 'audio': None, 'function_call': None, 'tool_calls': [], 'reasoning_content': None}, 'logprobs': None, 'finish_reason': 'stop', 'stop_reason': None, 'token_ids': None}], 'service_tier': None, 'system_fingerprint': None, 'usage': {'prompt_tokens': 646, 'total_tokens': 917, 'completion_tokens': 271, 'prompt_tokens_details': None}, 'prompt_logprobs': None, 'prompt_token_ids': None, 'kv_transfer_params': None}
+```
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/index.rst b/docs/vllm_mindspore/docs/source_zh_cn/index.rst
index d6ab0f650d82145db59d364c9398b0d06218b6ae..5f1a25db41bf97fb834192a52b9b066bbd420cdd 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/index.rst
+++ b/docs/vllm_mindspore/docs/source_zh_cn/index.rst
@@ -39,8 +39,8 @@ vLLM-MindSpore插件采用vLLM社区推荐的插件机制，实现能力注册
 - 操作系统：openEuler或Ubuntu Linux
 - 软件：
 
-  * Python >= 3.10, <= 3.12
-  * CANN >= 8.3.RC1
+  * Python >= 3.9, <= 3.11
+  * CANN == 8.3.RC1
   * MindSpore（与vLLM-MindSpore插件版本配套）
   * vLLM（与vLLM-MindSpore插件版本配套）
 
@@ -88,7 +88,7 @@ vLLM-MindSpore插件代码仓包含主干分支、开发分支、版本分支：
    *  -  r0.4.0
       -  Maintained
       -  基于vLLM v0.9.1版本和MindSpore 2.7.1版本CI看护
-   *  -  r0.5.0
+   *  -  v0.5.1
       -  Maintained
       -  基于vLLM v0.11.0版本和MindSpore 2.7.1.post1版本CI看护
 
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/release_notes/release_notes.md b/docs/vllm_mindspore/docs/source_zh_cn/release_notes/release_notes.md
index 8e628d26c73c21d213592f8ecd117cd86edc6bca..ddd45a05a96d3fe94a78cd42c9cd1c7b3225a926 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/release_notes/release_notes.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/release_notes/release_notes.md
@@ -2,9 +2,9 @@
 
 [![查看源文件](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.7.1.post1/resource/_static/logo_source.svg)](https://gitee.com/mindspore/docs/blob/r2.7.1.post1/docs/vllm_mindspore/docs/source_zh_cn/release_notes/release_notes.md)
 
-## vLLM-MindSpore插件 0.5.0 Release Notes
+## vLLM-MindSpore插件 0.5.1 Release Notes
 
-vLLM MindSpore插件0.5.0版本，配套vLLM 0.11.0版本。以下为此版本支持的关键新功能和模型。
+vLLM MindSpore插件0.5.1版本，配套vLLM 0.11.0版本。以下为此版本支持的关键新功能和模型。
 
 ### 新特性
 
@@ -18,6 +18,11 @@ vLLM MindSpore插件0.5.0版本，配套vLLM 0.11.0版本。以下为此版本
 - GLM系列模型
     - GLM-4.1V支持原生模型
 
+### 问题修复
+- 关于encoder的显存泄露问题
+    - [对于vLLM的修复](https://github.com/vllm-project/vllm/pull/31857)
+    - [对于vLLM-MindSpore插件的修复](https://gitee.com/mindspore/vllm-mindspore/pulls/1447)
+
 ### 贡献者
 
 感谢以下人员做出的贡献：
diff --git a/docs/vllm_mindspore/docs/source_zh_cn/user_guide/supported_models/models_list/models_list.md b/docs/vllm_mindspore/docs/source_zh_cn/user_guide/supported_models/models_list/models_list.md
index d50bf776cf935f65daa2da6a86729ade566ea3f6..cc9ee7860a86d821710745396d6107f2a883e51d 100644
--- a/docs/vllm_mindspore/docs/source_zh_cn/user_guide/supported_models/models_list/models_list.md
+++ b/docs/vllm_mindspore/docs/source_zh_cn/user_guide/supported_models/models_list/models_list.md
@@ -10,7 +10,7 @@
 | DeepSeek-R1 W8A4 |   已支持 | MindFormers | Atlas 800I A2 |  [DeepSeek-R1-W8A4](https://modelers.cn/models/MindSpore-Lab/DeepSeek-R1-0528-gs-A8W4) |
 | Telechat2 | 已支持 | MindFormers | Atlas 800I A2 | [TeleChat2-7B-32K](https://www.modelscope.cn/models/TeleAI/TeleChat2-7B-32K)、[TeleChat2-35B-32K](https://www.modelscope.cn/models/TeleAI/TeleChat2-35B-32K) |
 | GLM-4.5 | 已支持 | MindFormers | Atlas 800I A2 | [GLM-4.5](https://huggingface.co/zai-org/GLM-4.5), [GLM-4.5-Air](https://huggingface.co/zai-org/GLM-4.5-Air) |
-| GLM-4.1V-Thinking | 已支持 | MindFormers | Atlas 800I A2 | [GLM-4.1V-9B-Thinking](https://huggingface.co/zai-org/GLM-4.1V-9B-Thinking) |
+| GLM-4.1V-Thinking | 已支持 | Native | Atlas 800I A2 | [GLM-4.1V-9B-Thinking](https://huggingface.co/zai-org/GLM-4.1V-9B-Thinking) |
 | GLM-4 | 已支持 | MindFormers | Atlas 800I A2 | [GLM-4-9B-0414](https://huggingface.co/zai-org/GLM-4-9B-0414)、[GLM-4-32B-0414](https://huggingface.co/zai-org/GLM-4-32B-0414) |
 | Qwen3-VL | 已支持 | Native  | Atlas 800I A2 | [
 Qwen3-VL-2B](https://huggingface.co/Qwen/Qwen3-VL-2B-Instruct)、[Qwen3-VL-4B](https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct)、[Qwen3-VL-8B](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)、[Qwen3-VL-30B-A3B](https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct)、[Qwen3-VL-235B-A22B](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Instruct) |