Reduce tokens and change model

app.py

@@ -34,7 +34,7 @@ def answer_question(query):
     docs = load_raw_dataset()
     rag = CustomRAG(
         create_vector_database(docs, "all-MiniLM-L6-v2"),
-        get_llm("FreedomIntelligence/HuatuoGPT-o1-7B"),
+        get_llm("mistralai/Mistral-7B-Instruct-v0.1"),
         get_prompt_template(),
     )
     response, _ = rag.run(query)
@@ -52,7 +52,7 @@ demo = gr.Interface(
         ),
     ],
     outputs="text",
-    title="Medical Assistant – Powered by AI & RAG",
+    title="Medical Assistant – RAG",
     description=(
         "Get helpful insights based on your described symptoms. "
         "This assistant uses medical reference data to provide informative responses. "

indexer.py

@@ -42,7 +42,7 @@ def get_llm(model_name):
         "text-generation",
         model=model,
         tokenizer=tokenizer,
-        max_new_tokens=512,
+        max_new_tokens=300,
         temperature=0.7,
         do_sample=True,
     )