diff --git a/README.md b/README.md
index f08c1fc..55b01e8 100644
--- a/README.md
+++ b/README.md
@@ -1,10 +1,10 @@
 # Serverless Tiny Language Models
 
-[TinyLlama 4-bit quantized 3 trillion token chat model](https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF) and [8-bit quantized Qwen 2 beta 0.5B Chat](https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat-GGUF) running on Azure Functions consumption plan.
+[TinyLlama 1.1B 4-bit quantized 3 trillion token chat model](https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF) and [Qwen 2 beta 0.5B 8-bit quantized chat model](https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat-GGUF) running on Azure Functions consumption plan.
 
-This project is not intended for production use. It is a technology demonstration to show that it is possible to run a large language model on a cheap and scalable serverless platform.
+This project is not intended for production use. It is a technology demonstration to show that it is possible to run large language models on a cheap and scalable serverless platform.
 
-A demo of the app is available at [https://.azurewebsites.net/](https://.azurewebsites.net/). In the demo, you can enter a prompt and the model will generate a completion.
+A demo of the app is available at [https://tiny-serverless-llms.azurewebsites.net](https://tiny-serverless-llms.azurewebsites.net). In the demo, you can enter a prompt and the model will generate a completion.
 
 **Any abuse of the service will result in the service being taken down.**
 
@@ -39,13 +39,15 @@ You'll need the following resources in Azure:
 
 You'll need to set the following application setting in Azure Functions:
 
-`POST_BUILD_SCRIPT_PATH=post_build.sh`
-`MODEL_BASE=/home/site/wwwroot/`
-`LLAMA_BASE=/home/site/wwwroot/llama.cpp`
-`AzureSignalRBase=https://{signal_r_service_name}.service.signalr.net`
-`AzureSignalRAccessKey={signal_r_service_key}`
+```bash
+POST_BUILD_SCRIPT_PATH=post_build.sh
+MODEL_BASE=/home/site/wwwroot/
+LLAMA_BASE=/home/site/wwwroot/llama.cpp/
+AzureSignalRBase=https://{signal_r_service_name}.service.signalr.net
+AzureSignalRAccessKey={signal_r_service_key}
+```
 
-This script will be executed during Oryx build. It is used to build the llama.cpp binary and add the models to the deployment package.
+The `post_build.sh` script will be executed during Oryx build. It is used to build the llama.cpp binary and add the models to the deployment package.
 
 ## About [Softlandia](https://softlandia.fi/)