Update README.md

README.md

@@ -17,27 +17,10 @@ language:
 
 # DEMO https://youtu.be/CyHbX13AuK0
 
-> **Build agents that don't block the main thread**  
-> Stop waiting on responses. Add context anytime.  
-> Non-blocking, real-time AI.
-
-**Finetuned from model:** unsloth/Qwen3-14B-unsloth-bnb-4bit
+**Finetuned from model:** unsloth/Qwen3-14B-unsloth-bnb-4bit using [AsyncTool dataset](https://huggingface.co/datasets/qforge/AsyncTool)
 
 ![conversation](./image.png)
 
-## 🚀 What Makes This Model Special?
-
-This model has been fine-tuned to handle **asynchronous tool execution** — a critical capability for building responsive, real-world AI agents. Unlike traditional function-calling models that assume tools return results immediately, this model understands and properly handles tools that take time to execute.
-
-## 🔄 Async Tool Call Protocol
-
-The model implements a robust async protocol:
-
-1. **Tool Call**: The model makes a function/tool call
-2. **ACK (Acknowledgment)**: The tool immediately returns `<tool_ack id="tN"/>` to confirm the request is received
-3. **Processing**: The tool executes asynchronously (could be API calls, database queries, external services)
-4. **RESPONSE**: The tool returns the actual result later
-
 ## 💡 Why Async Tools?
 
 Real-world AI agents often need to:
@@ -48,7 +31,16 @@ Real-world AI agents often need to:
 - Handle multiple tool calls in parallel
 - Provide responsive user experiences without blocking
 
-Traditional function-calling models assume synchronous execution, leading to poor user experiences when tools take time to respond. This model solves that problem.
+This model handles **asynchronous tool execution** — a critical capability for building responsive, real-world AI agents. Unlike traditional function-calling models that assume tools return results immediately, this model understands and properly handles tools that take time to execute and return results later during conversation.
+
+## 🔄 Async Tool Call Protocol
+
+The model implements a robust async protocol:
+
+1. **Tool Call**: The model makes a function/tool call
+2. **ACK (Acknowledgment)**: The tool immediately returns `<tool_ack id="tN"/>` to confirm the request is received
+3. **Processing**: The tool executes asynchronously (could be API calls, database queries, external services)
+4. **RESPONSE**: The tool returns the actual result later
 
 ## 📋 Example Conversation Flow
 
@@ -108,7 +100,7 @@ https://youtu.be/CyHbX13AuK0
 
 ### Gemini
 
-We used Gemini Speech to Text and are currently fine tuning Gemini 2.5 Flash Lite Model.
+We used Gemini Speech to Text and are currently fine tuning Gemini 2.5 Flash Lite Model for the same task, improved latency and accuracy.
 
 ![Google Cloud](./google-cloud.png)
 
@@ -118,7 +110,10 @@ Link to implementation in Pipecat [Pull Request](https://github.com/pipecat-ai/p
 
 ## 5. Tell us what you did new during the hackathon
 
-At the hackathon we've improved our prepared earlier [dataset](https://huggingface.co/datasets/qforge/AsyncTool), fine tuned **unsloth/Qwen3-14B-unsloth-bnb-4bit** model (for handling Async Tools) and contributed to Pipecat by adding support for our new model ([Pull Request](https://github.com/pipecat-ai/pipecat/pull/2839)).
+At the hackathon we've 
+- improved [AsyncTool dataset](https://huggingface.co/datasets/qforge/AsyncTool) with more variety to improve quality of responses
+- fine tuned **unsloth/Qwen3-14B-unsloth-bnb-4bit** model using [Google Colab](https://colab.research.google.com/drive/1r6vSiTPODsN20NzdcfV58NWsu-kbcN-_) (for handling Async Tools) 
+- Prepared a draft ([Pull Request](https://github.com/pipecat-ai/pipecat/pull/2839)) to Pipecat by adding support for our new model and native behaviour
 
 ## 🔧 Training Details
 
@@ -128,4 +123,5 @@ This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unsloth
 
 #### 6. Feedback
 
-At the beginning we haved issues with running fine tuning on Google Vertex-ai (because of outdated documentation).
+At the beginning we had issues with running fine tuning on Google Vertex-ai (because of outdated documentation).
+Loved the test coverage and dev environment of pipecat.