Update app.py

app.py

@@ -41,14 +41,14 @@ def load_model():
         repo_id="Mykes/med_phi3-mini-4k-GGUF",
         filename="*Q4_K_M.gguf",
         verbose=False,
-        n_ctx=256,  # Reduced context window
-        n_batch=8,  # Smaller batch size
-        n_threads=2,  # Adjust based on your CPU cores
+        n_ctx=512,
+        n_batch=512,
+        n_threads=8,
+        use_mlock=True,
         use_mmap=True,
-        # n_ctx=1024,
-        # n_batch=512,
-        # n_threads=8,
-        # use_mlock=True,
+        # n_ctx=256,  # Reduced context window
+        # n_batch=8,  # Smaller batch size
+        # n_threads=2,  # Adjust based on your CPU cores
         # use_mmap=True,
     )
 
@@ -110,7 +110,7 @@ if prompt := st.chat_input("What is your question?"):
     st.session_state.messages.append({"role": "user", "content": prompt})
 
     # Format the context with the last 5 messages
-    context = format_context(st.session_state.messages[-5:])
+    context = format_context(st.session_state.messages[-3:])
     
     # Prepare the model input
     model_input = f"{context}Human: {prompt}\nAssistant:"