feat: using a larger model on rag agent

agents/BookRetriverAgent.py

@@ -1,4 +1,7 @@
 import os
+
+from langgraph.checkpoint.memory import InMemorySaver
+
 from tools import get_retrieve_book_context_tool
 from langchain.agents import create_agent
 from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
@@ -13,7 +16,7 @@ def get_book_retriver_agent():
 
 def generate_agent(vector_store):
     hf_model = HuggingFaceEndpoint(
-        repo_id="Qwen/Qwen3-VL-8B-Instruct",
+        repo_id="Qwen/Qwen3-30B-A3B-Instruct-2507",
         task="text-generation",
         provider="auto",
         huggingfacehub_api_token=os.getenv("HF_TOKEN")
@@ -24,13 +27,12 @@ def generate_agent(vector_store):
     prompt = """
         You are a knowledge retriever agent, you must always provide context related answers, using the tools provided.
         You must always use this tool for reliable information to answer any query.
-        Always prioritize data coming from your tools.
-        Dont use the same tool more than once.
+        Dont try to elaborate your answers, always prioritize data coming from your tools.
     """
     return create_agent(
         model=llm,
         tools=tools,
-        system_prompt=prompt
+        system_prompt=prompt,
     )
 
 def initiate_vector_store():

app.py

@@ -43,11 +43,9 @@ class GradioAgent:
         ]
 
         prompt = (
-            "You are a smart and friendly coordinator agent. "
-            "You have access to a collection of tools and specialized agents that help you provide accurate and reliable information. "
-            "You always communicate clearly, in a friendly and engaging tone — using short paragraphs, smooth transitions, and light emoji to make your answers pleasant and easy to read. "
-            "Keep the technical accuracy of your responses, but present them in a natural, conversational way. "
-            "Your main goal is to generate user-friendly and informative answers based on the data you gather."
+            "You are a helpful and usefull coordinator agent, you have access to a collection of tools and"
+            " agents to help you with reliable data to your query's. "
+            "One of your main objectives is to generate user friendly answers based on the information you have."
         )
 
         return create_agent(