diff --git a/litgpt/deploy/serve.py b/litgpt/deploy/serve.py
index 44aa0b57..bab53246 100644
--- a/litgpt/deploy/serve.py
+++ b/litgpt/deploy/serve.py
@@ -233,7 +233,9 @@ def run_server(
             The "auto" setting (default) chooses a GPU if available, and otherwise uses a CPU.
         port: The network port number on which the model is configured to be served.
         stream: Whether to stream the responses.
-        openai_spec: Whether to use the OpenAISpec.
+        openai_spec: Whether to use the OpenAISpec and enable OpenAI-compatible API endpoints. When True, the server will provide
+            `/v1/chat/completions` endpoints that work with the OpenAI SDK and other OpenAI-compatible clients,
+            making it easy to integrate with existing applications that use the OpenAI API.
         access_token: Optional API token to access models with restrictions.
     """
     checkpoint_dir = auto_download_checkpoint(model_name=checkpoint_dir, access_token=access_token)
diff --git a/tutorials/deploy.md b/tutorials/deploy.md
index cd129bbe..b58a08ac 100644
--- a/tutorials/deploy.md
+++ b/tutorials/deploy.md
@@ -80,6 +80,60 @@ Sure, here is the corrected sentence:
 Example input
 ```
 
+ 
+## Serve an LLM with OpenAI-compatible API
+
+LitGPT provides OpenAI-compatible endpoints that allow you to use the OpenAI SDK or any OpenAI-compatible client to interact with your models. This is useful for integrating LitGPT into existing applications that use the OpenAI API.
+
+ 
+### Step 1: Start the server with OpenAI specification
+
+```bash
+# 1) Download a pretrained model (alternatively, use your own finetuned model)
+litgpt download HuggingFaceTB/SmolLM2-135M-Instruct
+
+# 2) Start the server with OpenAI-compatible endpoints
+litgpt serve HuggingFaceTB/SmolLM2-135M-Instruct --openai_spec true
+```
+
+> [!TIP]
+> The `--openai_spec true` flag enables OpenAI-compatible endpoints at `/v1/chat/completions` instead of the default `/predict` endpoint.
+
+ 
+### Step 2: Query using OpenAI-compatible endpoints
+
+You can now send requests to the OpenAI-compatible endpoint using curl:
+
+```bash
+curl -X POST http://127.0.0.1:8000/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "SmolLM2-135M-Instruct",
+    "messages": [{"role": "user", "content": "Hello! How are you?"}]
+  }'
+```
+
+Or use the OpenAI Python SDK:
+
+```python
+from openai import OpenAI
+
+# Configure the client to use your local LitGPT server
+client = OpenAI(
+    base_url="http://127.0.0.1:8000/v1",
+    api_key="not-needed"  # LitGPT doesn't require authentication by default
+)
+
+response = client.chat.completions.create(
+    model="SmolLM2-135M-Instruct",
+    messages=[
+        {"role": "user", "content": "Hello! How are you?"}
+    ]
+)
+
+print(response.choices[0].message.content)
+```
+
  
 ## Serve an LLM UI with Chainlit