Merge pull request #99 from alipay/dev_deepseek

add deepseek doc

docs/guidebook/en/3_1_2_DeepSeek_LLM_Use.md

@@ -0,0 +1,35 @@
+# DeepSeek  Usage
+## 1. Create the relevant file.
+Create a YAML file, for example, user_deepseek_llm.yaml
+Paste the following content into your user_deepseek_llm.yaml file.
+```yaml
+name: 'user_deepseek_llm'
+description: 'default default_deepseek_llm llm with spi'
+model_name: 'deepseek-chat'
+max_tokens: 1000
+metadata:
+  type: 'LLM'
+  module: 'agentuniverse.llm.default.deep_seek_openai_style_llm'
+  class: 'DefaultDeepSeekLLM'
+```
+##  2. Environment Setup
+Must be configured: DEEPSEEK_API_KEY  
+Optional: DEEPSEEK_API_BASE
+### 2.1 Configure through Python code
+```python
+import os
+os.environ['DEEPSEEK_API_KEY'] = 'sk-***'
+os.environ['DEEPSEEK_API_BASE'] = 'https://xxxxxx'
+```
+### 2.2 Configure through the configuration file
+In the custom_key.toml file under the config directory of the project, add the configuration:
+```toml
+DEEPSEEK_API_KEY="sk-******"
+DEEPSEEK_API_BASE="https://xxxxxx"
+```
+## 3. Obtaining the DeepSeek API KEY 
+Please refer to the official documentation of DEEPSEEK:https://platform.deepseek.com/api_keys
+
+## 4. Tips
+In the agentuniverse, we have established an LLM named default_deepseek_llm. Users can directly utilize it after configuring their DEEPSEEK_API_KEY.
+

docs/guidebook/en/3_1_2_OpenAIStyleLLM_Use.md

@@ -0,0 +1,37 @@
+# OpenAIStyleLLM Usage
+## 1. Create the relevant file.
+Create a YAML file, for example, user_openai_style_llm.yaml
+Paste the following content into your user_openai_style_llm.yaml file.
+```yaml
+name: 'deep_seek_llm'
+description: 'demo deep_seek llm with spi'
+model_name: 'deepseek-chat'
+max_tokens: 4000
+max_context_length: 32000
+api_key_env: 'DEEPSEEK_API_KEY'
+api_base_env: 'DEEPSEEK_API_BASE'
+metadata:
+  type: 'LLM'
+  module: 'agentuniverse.llm.openai_style_llm'
+  class: 'OpenAIStyleLLM'
+```
+Parameter Description:
+max_context_length: The context length that the model can handle.
+api_key_env: The name of the environment variable for the API key.
+api_base_env: The name of the environment variable for the API base.
+These parameters must be configured, and upon service startup, the corresponding values will be loaded from the environment variables.
+##  2. Environment Setup
+Mandatory Configuration: $api_key_env, $api_base_env.
+In the configuration file, api_key_env and api_base_env are set as DEEPSEEK_API_KEY and DEEPSEEK_API_BASE respectively. Therefore, the configuration should be as follows:
+### 2.1 Configure through Python code
+```python
+import os
+os.environ['DEEPSEEK_API_KEY'] = 'sk-***'
+os.environ['DEEPSEEK_API_BASE'] = 'https://xxxxxx'
+```
+### 2.2 Configure through the configuration file
+In the custom_key.toml file under the config directory of the project, add the configuration:
+```toml
+DEEPSEEK_API_KEY="DEEPSEEK_API_KEY"
+DEEPSEEK_API_BASE="https://xxxxxx"
+```

docs/guidebook/zh/3_1_2_DeepSeek使用.md

@@ -0,0 +1,35 @@
+# DeepSeek 使用
+## 1. 创建相关文件
+创建一个yaml文件，例如 user_deepseek_llm.yaml
+将以下内容粘贴到您的user_deepseek_llm.yaml文件当中
+```yaml
+name: 'user_deepseek_llm'
+description: 'default default_deepseek_llm llm with spi'
+model_name: 'deepseek-chat'
+max_tokens: 1000
+metadata:
+  type: 'LLM'
+  module: 'agentuniverse.llm.default.deep_seek_openai_style_llm'
+  class: 'DefaultDeepSeekLLM'
+```
+## 2. 环境设置
+### 2.1 通过python代码配置
+必须配置：DEEPSEEK_API_KEY  
+可选配置：DEEPSEEK_API_BASE
+```python
+import os
+os.environ['DEEPSEEK_API_KEY'] = 'sk-***'
+os.environ['DEEPSEEK_API_BASE'] = 'https://xxxxxx'
+```
+### 2.2 通过配置文件配置
+在项目的config目录下的custom_key.toml当中，添加配置：
+```toml
+DEEPSEEK_API_KEY="sk-******"
+DEEPSEEK_API_BASE="https://xxxxxx"
+```
+## 3. DEEPSEEK API KEY 获取
+参考 DEEPSEEK 官方文档：https://platform.deepseek.com/api_keys
+
+## 4. Tips
+在agentuniverse中，我们已经创建了一个name为default_deepseek_llm的llm,用户在配置DEEPSEEK_API_KEY之后可以直接使用。
+

docs/guidebook/zh/3_1_2_OpenAIStyleLLM使用.md

@@ -0,0 +1,39 @@
+# OpenAIStyleLLM 使用
+通过该配置，您可以连接任何遵守OpenAI标准的模型服务
+## 1. 创建相关文件
+创建一个yaml文件，例如 user_openai_style_llm.yaml
+将以下内容粘贴到您的user_openai_style_llm.yaml文件当中
+```yaml
+name: 'deep_seek_llm'
+description: 'demo deep_seek llm with spi'
+model_name: 'deepseek-chat'
+max_tokens: 4000
+max_context_length: 32000
+api_key_env: 'DEEPSEEK_API_KEY'
+api_base_env: 'DEEPSEEK_API_BASE'
+metadata:
+  type: 'LLM'
+  module: 'agentuniverse.llm.openai_style_llm'
+  class: 'OpenAIStyleLLM'
+```
+参数说明：
+    max_context_length: 模型所能承接的上下文
+    api_key_env: api key的env变量名
+    api_base_env: api base的env变量名
+上述参数必须配置,服务启动时会从环境变量中加载相关的值
+## 2. 环境设置
+必须配置：$api_key_env, $api_base_env
+配置文件中api_key_env与api_base_env分别为DEEPSEEK_API_KEY、DEEPSEEK_API_BASE，所以配置如下：
+### 2.1 通过python代码配置
+```python
+import os
+os.environ['DEEPSEEK_API_KEY'] = 'sk-***'
+os.environ['DEEPSEEK_API_BASE'] = 'https://xxxxxx'
+```
+### 2.2 通过配置文件配置
+在项目的config目录下的custom_key.toml当中，添加配置：
+```toml
+DEEPSEEK_API_KEY="DEEPSEEK_API_KEY"
+DEEPSEEK_API_BASE="https://xxxxxx"
+```
+注意，这里必须与配置文件中锁配置的api_key_env、api_base_env一致