update deepseek docs

Signed-off-by: oilbeater <[email protected]>

README.md

@@ -68,9 +68,13 @@ It is written in **TypeScript**: ensuring adaptability to the rapidly evolving A
 
    Then you can visit Malacca with the Worker domain or custom domain.
 
-### How to use?
+### Supported LLM Providers
 
 - [Azure OpenAI](./docs/azure_openai.md)
+- [DeepSeek](./docs/deepseek.md) 
+
+### How to use?
+
 - [Virtual Key](./docs/virtual-key.md)
 - [Caching](./docs/caching.md)
 - [Guard](./docs/guards.md)

docs/azure_openai.md

@@ -1,6 +1,6 @@
-# Azure OpenAI Usage
+# Azure OpenAI
 
-To leverage the advanced features provided by Malacca while using the Azure OpenAI API, you only need to adjust the `base_url` configuration in their SDK. This simple change allows you to benefit from Malacca's capabilities such as caching, virtual key management, rate limiting, and logging.
+To leverage the advanced features provided by Malacca while using the Azure OpenAI API, you only need to adjust the `base_url` configuration in the original SDK. This simple change allows you to benefit from Malacca's capabilities such as caching, virtual key management, rate limiting, logging and all other features.
 
 You need to get the following parameters:
 

docs/deepseek.md

@@ -0,0 +1,29 @@
+# DeepSeek
+
+To leverage the advanced features provided by Malacca while using the DeepSeek API, you only need to adjust the `base_url` configuration in the original SDK. This simple change allows you to benefit from Malacca's capabilities such as caching, virtual key management, rate limiting, logging and all other features.
+
+You need to get the following parameters:
+
+- `Worker_Domain`: The domain url where your Worker is accessible, you can find it in your worker Settings.
+- `Virtual_Key`: The virtual key that protect your real API key.
+
+## Python Example
+
+DeepSeek API is compatible with OpenAI API, so you can use [OpenAI Python SDK](https://github.com/openai/openai-python) to visit DeepSeek with Malacca endpoint.
+
+```python
+from openai import OpenAI
+
+client = OpenAI(api_key="{Virtual_Key}", base_url="https://{Worker_Domain}/deepseek/")
+
+response = client.chat.completions.create(
+    model="deepseek-chat",
+    messages=[
+        {"role": "system", "content": "You are a helpful assistant"},
+        {"role": "user", "content": "Tell me a very short story about Malacca."},
+    ],
+    stream=False
+)
+
+print(response.choices[0].message.content)
+```