diff --git a/README.md b/README.md
index c84f24330..d456aaef3 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 # Ollama
 
-Ollama is a tool for running any large language model on any machine. It's designed to be easy to use and fast, supporting the largest number of models possible by using the fastest loader available for your platform and model.
+Ollama is a tool for running large language models. It's designed to be easy to use and fast.
 
 > _Note: this project is a work in progress. Certain models that can be run with `ollama` are intended for research and/or non-commercial use only._
 
@@ -38,63 +38,7 @@ Or directly via downloaded model files:
 ollama run ~/Downloads/orca-mini-13b.ggmlv3.q4_0.bin
 ```
 
-## Python SDK
-
-### Example
-
-```python
-import ollama
-ollama.generate("orca-mini-3b", "hi")
-```
-
-### `ollama.generate(model, message)`
-
-Generate a completion
-
-```python
-ollama.generate("./llama-7b-ggml.bin", "hi")
-```
-
-### `ollama.models()`
-
-List available local models
-
-```python
-models = ollama.models()
-```
-
-### `ollama.load(model)`
-
-Manually a model for generation
-
-```python
-ollama.load("model")
-```
-
-### `ollama.unload(model)`
-
-Unload a model
-
-```python
-ollama.unload("model")
-```
-
-### `ollama.pull(model)`
-
-Download a model
-
-```python
-ollama.pull("huggingface.co/thebloke/llama-7b-ggml")
-```
-
-### `ollama.search(query)`
-
-Search for compatible models that Ollama can run
-
-```python
-ollama.search("llama-7b")
-```
-
 ## Documentation
 
 - [Development](docs/development.md)
+- [Python SDK](docs/python.md)
diff --git a/docs/python.md b/docs/python.md
new file mode 100644
index 000000000..04732ccf6
--- /dev/null
+++ b/docs/python.md
@@ -0,0 +1,64 @@
+# Python SDK
+
+## Install
+
+```
+pip install ollama
+```
+
+## Example
+
+```python
+import ollama
+ollama.generate("orca-mini-3b", "hi")
+```
+
+## Reference
+
+### `ollama.generate(model, message)`
+
+Generate a completion
+
+```python
+ollama.generate("./llama-7b-ggml.bin", "hi")
+```
+
+### `ollama.models()`
+
+List available local models
+
+```python
+models = ollama.models()
+```
+
+### `ollama.load(model)`
+
+Manually a model for generation
+
+```python
+ollama.load("model")
+```
+
+### `ollama.unload(model)`
+
+Unload a model
+
+```python
+ollama.unload("model")
+```
+
+### `ollama.pull(model)`
+
+Download a model
+
+```python
+ollama.pull("huggingface.co/thebloke/llama-7b-ggml")
+```
+
+### `ollama.search(query)`
+
+Search for compatible models that Ollama can run
+
+```python
+ollama.search("llama-7b")
+```