diff --git a/README.md b/README.md
index f5bae0b..1775020 100644
--- a/README.md
+++ b/README.md
@@ -3,6 +3,8 @@
 
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE) [![CI](https://img.shields.io/badge/ci-passing-brightgreen)](<LINK_TO_CI>)
 
+📄 **Read the paper:** [fastWorkflow: Closing the Performance Gap Between Small and Frontier Language Models for Conversational Agents](https://github.com/radiantlogicinc/fastworkflow/blob/main/paper/fastWorkflow_Closing_the_Performance_Gap_Between_Small_and_Frontier_Language_Models_for_Conversational_Agents.pdf)
+
 Build agents and assistants for complex workflows and large-scale Python applications, with deterministic or AI-powered business logic.
 
 If you have tried AI enabling non-trivial applications, you have struggled with the following:
@@ -23,10 +25,10 @@ While [DSPy](https://dspy.ai) ([Why DSPy](https://x.com/lateinteraction/status/1
   <table>
     <tr>
       <td align="center" width="50%">
-        <img src="fastWorkflow - Tau Bench Retail.jpg" alt="fastWorkflow - Tau Bench Retail" style="max-width: 100%; height: auto;"/>
+        <img src="tau_bench_retail_performance.png" alt="Tau-Bench Retail Performance" style="max-width: 100%; height: auto;"/>
       </td>
       <td align="center" width="50%">
-        <img src="fastWorkflow - TauBench Airline.jpg" alt="fastWorkflow - TauBench Airline" style="max-width: 100%; height: auto;"/>
+        <img src="tau_bench_airline_performance.png" alt="Tau-Bench Airline Performance" style="max-width: 100%; height: auto;"/>
       </td>
     </tr>
   </table>
diff --git a/fastWorkflow - Tau Bench Retail.jpg b/fastWorkflow - Tau Bench Retail.jpg
deleted file mode 100644
index a65fac6..0000000
Binary files a/fastWorkflow - Tau Bench Retail.jpg and /dev/null differ
diff --git a/fastWorkflow - TauBench Airline.jpg b/fastWorkflow - TauBench Airline.jpg
deleted file mode 100644
index 8b8b13f..0000000
Binary files a/fastWorkflow - TauBench Airline.jpg and /dev/null differ
diff --git a/paper/fastWorkflow_Closing_the_Performance_Gap_Between_Small_and_Frontier_Language_Models_for_Conversational_Agents.pdf b/paper/fastWorkflow_Closing_the_Performance_Gap_Between_Small_and_Frontier_Language_Models_for_Conversational_Agents.pdf
new file mode 100644
index 0000000..1ad6bd5
Binary files /dev/null and b/paper/fastWorkflow_Closing_the_Performance_Gap_Between_Small_and_Frontier_Language_Models_for_Conversational_Agents.pdf differ
diff --git a/tau_bench_airline_performance.png b/tau_bench_airline_performance.png
new file mode 100644
index 0000000..5b4273e
Binary files /dev/null and b/tau_bench_airline_performance.png differ
diff --git a/tau_bench_retail_performance.png b/tau_bench_retail_performance.png
new file mode 100644
index 0000000..6be0026
Binary files /dev/null and b/tau_bench_retail_performance.png differ