initial commit

streamlit_app.py

@@ -0,0 +1,49 @@
+from llama_cpp import Llama
+import streamlit as st
+
+model_path = "vicuna-13b-v1.5.ggmlv3.q2_K.bin"
+llama = Llama(model_path)
+
+def generate_response(messages: list) -> str:
+    response = llama.create_chat_completion(messages, max_tokens=-1, stream=False)
+    print(f"response: {response}")
+    return response['choices'][0]['message']['content']
+
+def main():
+    st.title("Chat with Vicuna!")
+
+    # Session state for retaining messages
+    if 'messages' not in st.session_state:
+        st.session_state.messages = []
+
+    # Display chat messages from history on app rerun
+    for message in st.session_state.messages:
+        with st.chat_message(message["role"]):
+            st.markdown(f"{message['content']}")
+
+    # Input for the user message
+    user_message = st.chat_input("Your Message")
+
+    # React to user input
+    if user_message:
+        # Display user message in chat message container
+        with st.chat_message("user"):
+            st.markdown(f"{user_message}")
+        # Add user message to chat history
+        st.session_state.messages.append({"role": "user", "content": user_message})
+
+        with st.chat_message("assistant"):
+            message_placeholder = st.empty()
+            full_response = ""
+
+            for char in generate_response([{"role": m["role"], "content": m["content"]} for m in st.session_state.messages]):
+                full_response += char
+                message_placeholder.markdown(full_response + "❙")
+
+            message_placeholder.markdown(full_response)
+
+        st.session_state.messages.append({"role": "assistant", "content": full_response})
+
+
+if __name__ == "__main__":
+    main()