Move installation section of readme higher. (#8)

Author: richwhitjr

README.md

@@ -2,10 +2,12 @@
 
 Joinly is a library designed join two different lists of keyed values using a large language model (LLM). Like a human data annotator, you can provide context through a prompt to specify how you want the task accomplished. The library supports inner, left, right, and full joins.
 
-By default, the library uses OpenAI's GPT-4o model. To use the library, set your OpenAI credentials as follows:
+# Installation
+
+Install Joinly using pip:
 
 ```
-OPENAI_API_KEY=XXXX
+pip install joinly
 ```
 
 # Format
@@ -23,14 +25,6 @@ right: List[Tuple[str, Any]] = []
 results: List[Tuple[Tuple[str, Any], Tuple[str, Any]]] = join.inner_join(left, right)
 ```
 
-# Installation
-
-Install Joinly using pip:
-
-```
-pip install joinly
-```
-
 # Quickstart
 
 Suppose you have two lists of fruits. One is more specific about the types of fruits, while the other is more general. Joinly allows you to describe the task and join the lists in a fuzzy manner using LLMs.
@@ -64,6 +58,12 @@ results = join.inner_join(left, right, context=context)
 ]
 ```
 
+By default, the library uses OpenAI's GPT-4o model. To use the library, set your OpenAI credentials as follows:
+
+```
+OPENAI_API_KEY=XXXX
+```
+
 # Custom LLMs
 
 You can integrate any custom model by implementing the `BaseAI` class with an embedding method and a prompt.