commit

server/main.py

@@ -58,10 +58,10 @@ async def lifespan(app:FastAPI):
     groq_api_key= os.getenv("groq_api_key")
     google_api_key= os.getenv("google_api_key")
     
-    primary_heavy_llm= ChatGroq(groq_api_key= groq_api_key,model_name="llama-3.3-70b-versatile",temperature=0.1,max_tokens=1024,timeout=10)
-    backup_heavy_llm1= ChatGroq(groq_api_key= groq_api_key,model_name="llama-3.1-8b-instant",temperature=0.1,max_tokens=1024,timeout=8)
-    backup_heavy_llm2= ChatGoogleGenerativeAI(google_api_key= google_api_key,model="models/gemini-3.5-flash",temperature=0.1,max_tokens=1024,timeout=8)
-    backup_heavy_llm3= ChatGoogleGenerativeAI(google_api_key= google_api_key,model="models/gemini-3.1-flash-lite",temperature=0.1,max_tokens=1024,timeout=8)
+    primary_heavy_llm= ChatGroq(groq_api_key= groq_api_key,model_name="llama-3.3-70b-versatile",temperature=0.1,max_tokens=1024,timeout=10,max_retries=0)
+    backup_heavy_llm1= ChatGroq(groq_api_key= groq_api_key,model_name="llama-3.1-8b-instant",temperature=0.1,max_tokens=1024,timeout=8,max_retries=0)
+    backup_heavy_llm2= ChatGoogleGenerativeAI(google_api_key= google_api_key,model="models/gemini-3.5-flash",temperature=0.1,max_tokens=1024,timeout=8,max_retries=0)
+    backup_heavy_llm3= ChatGoogleGenerativeAI(google_api_key= google_api_key,model="models/gemini-3.1-flash-lite",temperature=0.1,max_tokens=1024,timeout=8,max_retries=0)
     backup_heavy_llm4= ChatGoogleGenerativeAI(google_api_key= google_api_key,model="models/gemini-2.5-flash",temperature=0.1,max_tokens=1024,timeout=8)
 
     resilient_heavy_llm= primary_heavy_llm.with_fallbacks([backup_heavy_llm1,backup_heavy_llm2,backup_heavy_llm3,backup_heavy_llm4])

server/utils/RetrieveQuery.py

@@ -38,7 +38,7 @@ def QueryRetriever(queries,vector_retriever,keyword_retriever,chunks_dict):
     # sort on basis of values
     rrf_scores= sorted(rrf_scores.items(), key=lambda item: item[1], reverse=True)
     # select top 10 documents
-    for chunk_id,score in rrf_scores[:10]:
+    for chunk_id,score in rrf_scores[:5]:
         if chunk_id in chunks_dict:
             documents.append(final_documents[chunk_id])