diff --git a/README.md b/README.md
index e9cef35..6518e33 100644
--- a/README.md
+++ b/README.md
@@ -72,6 +72,8 @@ profiles:
     - "llama"
 ```
 
+More complex [examples](examples/README.md) for different use cases.
+
 ## Installation
 
 1. Create a configuration file, see [config.example.yaml](config.example.yaml)
diff --git a/examples/README.md b/examples/README.md
new file mode 100644
index 0000000..8e46a31
--- /dev/null
+++ b/examples/README.md
@@ -0,0 +1,9 @@
+# Example Configurations
+
+Learning by example is best.
+
+Here in the `examples/` folder are llama-swap configurations that can be used on your local LLM server.
+
+## List
+
+* [Speculative Decoding](speculative-decoding/README.md) - using a small draft model can increase inference speeds from 20% to 40%. This example includes a configurations Qwen2.5-Coder-32B (2.5x increase) and Llama-3.1-70B (1.4x increase) in the best cases.
diff --git a/examples/speculative-decoding/README.md b/examples/speculative-decoding/README.md
new file mode 100644
index 0000000..ff8340d
--- /dev/null
+++ b/examples/speculative-decoding/README.md
@@ -0,0 +1,3 @@
+# Qwen 2.5 Coder with a Draft Model
+
+Using a small draft model like qwen-2.5-coder-0.5B can have a big impact on the performance of the larger 32 billion parameter model.
\ No newline at end of file