example: add vlm to token in & out example (#3941)

Co-authored-by: zhaochenyang20 <zhaochen20@outlook.com>

examples/runtime/engine/readme.md

@@ -9,15 +9,15 @@ SGLang provides a direct inference engine without the need for an HTTP server. T
 
 ## Examples
 
-### 1. [Offline Batch Inference](./offline_batch_inference.py)
+### [Offline Batch Inference](./offline_batch_inference.py)
 
 In this example, we launch an SGLang engine and feed a batch of inputs for inference. If you provide a very large batch, the engine will intelligently schedule the requests to process efficiently and prevent OOM (Out of Memory) errors.
 
-### 2. [Embedding Generation](./embedding.py)
+### [Embedding Generation](./embedding.py)
 
 In this example, we launch an SGLang engine and feed a batch of inputs for embedding generation.
 
-### 3. [Custom Server](./custom_server.py)
+### [Custom Server](./custom_server.py)
 
 This example demonstrates how to create a custom server on top of the SGLang Engine. We use [Sanic](https://sanic.dev/en/) as an example. The server supports both non-streaming and streaming endpoints.
 
@@ -43,3 +43,7 @@ curl -X POST http://localhost:8000/generate_stream  -H "Content-Type: applicatio
 ```
 
 This will send both non-streaming and streaming requests to the server.
+
+### [Token-In-Token-Out for RLHF](./token_in_token_out)
+
+In this example, we launch an SGLang engine, feed tokens as input and generate tokens as output.