Bump version to v0.7.1 (#3178)

* bump version to v0.7.1

* update pipeline guide

* update

* fix according reviewer's comment

Author: lvhan028

README.md

@@ -23,6 +23,10 @@ ______________________________________________________________________
 
 ## Latest News 🎉
 
+<details open>
+<summary><b>2025</b></summary>
+</details>
+
 <details close>
 <summary><b>2024</b></summary>
 
@@ -195,9 +199,9 @@ For more information on installing on CUDA 11+ platform, or for instructions on
 
 ```python
 import lmdeploy
-pipe = lmdeploy.pipeline("internlm/internlm3-8b-instruct")
-response = pipe(["Hi, pls intro yourself", "Shanghai is"])
-print(response)
+with lmdeploy.pipeline("internlm/internlm3-8b-instruct") as pipe:
+    response = pipe(["Hi, pls intro yourself", "Shanghai is"])
+    print(response)
 ```
 
 > \[!NOTE\]

README_ja.md

@@ -23,6 +23,10 @@ ______________________________________________________________________
 
 ## 最新ニュース 🎉
 
+<details open>
+<summary><b>2025</b></summary>
+</details>
+
 <details close>
 <summary><b>2024</b></summary>
 
@@ -192,9 +196,9 @@ CUDA 11+プラットフォームでのインストールに関する情報、ま
 
 ```python
 import lmdeploy
-pipe = lmdeploy.pipeline("internlm/internlm3-8b-instruct")
-response = pipe(["Hi, pls intro yourself", "Shanghai is"])
-print(response)
+with lmdeploy.pipeline("internlm/internlm3-8b-instruct") as pipe:
+    response = pipe(["Hi, pls intro yourself", "Shanghai is"])
+    print(response)
 ```
 
 > \[!NOTE\]

README_zh-CN.md

@@ -23,6 +23,10 @@ ______________________________________________________________________
 
 ## 最新进展 🎉
 
+<details open>
+<summary><b>2025</b></summary>
+</details>
+
 <details close>
 <summary><b>2024</b></summary>
 
@@ -196,9 +200,9 @@ pip install lmdeploy
 
 ```python
 import lmdeploy
-pipe = lmdeploy.pipeline("internlm/internlm3-8b-instruct")
-response = pipe(["Hi, pls intro yourself", "Shanghai is"])
-print(response)
+with lmdeploy.pipeline("internlm/internlm3-8b-instruct") as pipe:
+    response = pipe(["Hi, pls intro yourself", "Shanghai is"])
+    print(response)
 ```
 
 > \[!NOTE\]

docs/en/get_started/installation.md

@@ -23,7 +23,7 @@ pip install lmdeploy
 The default prebuilt package is compiled on **CUDA 12**. If CUDA 11+ (>=11.3) is required, you can install lmdeploy by:
 
 ```shell
-export LMDEPLOY_VERSION=0.7.0.post3
+export LMDEPLOY_VERSION=0.7.1
 export PYTHON_VERSION=38
 pip install https://github.com/InternLM/lmdeploy/releases/download/v${LMDEPLOY_VERSION}/lmdeploy-${LMDEPLOY_VERSION}+cu118-cp${PYTHON_VERSION}-cp${PYTHON_VERSION}-manylinux2014_x86_64.whl --extra-index-url https://download.pytorch.org/whl/cu118
 ```

docs/en/llm/pipeline.md

@@ -217,6 +217,18 @@ response = pipe(prompts, gen_config=gen_config, adapter_name='lora_name_1')
 print(response)
 ```
 
+### Release pipeline
+
+You can release the pipeline explicitly by calling its `close()` method, or alternatively, use the `with` statement as demonstrated below:
+
+```python
+from lmdeploy import pipeline
+
+with pipeline('internlm/internlm2_5-7b-chat') as pipe:
+    response = pipe(['Hi, pls intro yourself', 'Shanghai is'])
+    print(response)
+```
+
 ## FAQs
 
 - **RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase**.

docs/en/multi_modal/vl_pipeline.md

@@ -213,3 +213,25 @@ print(sess.response.text)
 sess = pipe.chat('What is the woman doing?', session=sess, gen_config=gen_config)
 print(sess.response.text)
 ```
+
+## Release pipeline
+
+You can release the pipeline explicitly by calling its `close()` method, or alternatively, use the `with` statement as demonstrated below:
+
+```python
+from lmdeploy import pipeline
+
+from lmdeploy import pipeline
+from lmdeploy.vl import load_image
+
+with pipeline('OpenGVLab/InternVL2_5-8B') as pipe:
+    image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
+    response = pipe(('describe this image', image))
+    print(response)
+
+# Clear the torch cache and perform garbage collection if needed
+import torch
+import gc
+torch.cuda.empty_cache()
+gc.collect()
+```

docs/zh_cn/get_started/installation.md

@@ -23,7 +23,7 @@ pip install lmdeploy
 默认的预构建包是在 **CUDA 12** 上编译的。如果需要 CUDA 11+ (>=11.3)，你可以使用以下命令安装 lmdeploy：
 
 ```shell
-export LMDEPLOY_VERSION=0.7.0.post3
+export LMDEPLOY_VERSION=0.7.1
 export PYTHON_VERSION=38
 pip install https://github.com/InternLM/lmdeploy/releases/download/v${LMDEPLOY_VERSION}/lmdeploy-${LMDEPLOY_VERSION}+cu118-cp${PYTHON_VERSION}-cp${PYTHON_VERSION}-manylinux2014_x86_64.whl --extra-index-url https://download.pytorch.org/whl/cu118
 ```

docs/zh_cn/llm/pipeline.md

@@ -223,6 +223,18 @@ response = pipe(prompts, gen_config=gen_config, adapter_name='lora_name_1')
 print(response)
 ```
 
+### 释放 pipeline
+
+您可以通过调用其 `close()` 方法来显式释放 pipeline，或者，也可以使用 `with` 语句，如下所示：
+
+```python
+from lmdeploy import pipeline
+
+with pipeline('internlm/internlm2_5-7b-chat') as pipe:
+    response = pipe(['Hi, pls intro yourself', 'Shanghai is'])
+    print(response)
+```
+
 ## 常见问题
 
 - **RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase**.

docs/zh_cn/multi_modal/vl_pipeline.md

@@ -213,3 +213,25 @@ print(sess.response.text)
 sess = pipe.chat('What is the woman doing?', session=sess, gen_config=gen_config)
 print(sess.response.text)
 ```
+
+### 释放 pipeline
+
+您可以通过调用其 `close()` 方法来显式释放 pipeline，或者，也可以使用 `with` 语句，如下所示：
+
+```python
+from lmdeploy import pipeline
+
+from lmdeploy import pipeline
+from lmdeploy.vl import load_image
+
+with pipeline('OpenGVLab/InternVL2_5-8B') as pipe:
+    image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
+    response = pipe(('describe this image', image))
+    print(response)
+
+# Clear the torch cache and perform garbage collection if needed
+import torch
+import gc
+torch.cuda.empty_cache()
+gc.collect()
+```

lmdeploy/version.py

@@ -1,7 +1,7 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 from typing import Tuple
 
-__version__ = '0.7.0.post3'
+__version__ = '0.7.1'
 short_version = __version__