Update app.py

app.py

@@ -2,138 +2,89 @@ import streamlit as st
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 from threading import Thread
 import torch
-import time
 
-# 1. Advanced CSS: Branding, Animations, and Split-Pane Artifacts
-st.set_page_config(page_title="Claude", page_icon="☁️", layout="wide")
+# App Configuration
+st.set_page_config(page_title="Qwen3 Turbo", page_icon="⚡")
 
-# Claude's exact 2026 color palette
-CLAUDE_ORANGE = "#d97757"
-CLAUDE_PAPER = "#f9f9f8"
-
-st.markdown(f"""
-    <style>
-        @import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600&display=swap');
-        
-        .stApp {{ background-color: #ffffff; font-family: 'Inter', sans-serif; }}
-        
-        /* Sidebar: warm paper texture */
-        [data-testid="stSidebar"] {{
-            background-color: {CLAUDE_PAPER} !important;
-            border-right: 1px solid #e5e5e5 !important;
-        }}
-
-        /* The "Living Avatar" Animation - Moving Lines Effect */
-        @keyframes breathe {{
-            0% {{ transform: scale(1); opacity: 0.8; }}
-            50% {{ transform: scale(1.05); opacity: 1; }}
-            100% {{ transform: scale(1); opacity: 0.8; }}
-        }}
-        
-        [data-testid="chatAvatarAssistant"] {{
-            background-color: {CLAUDE_ORANGE} !important;
-            border-radius: 8px !important;
-            animation: breathe 3s infinite ease-in-out;
-            box-shadow: 0 0 15px rgba(217, 119, 87, 0.2);
-        }}
-
-        /* Artifacts Window Styling */
-        .artifact-container {{
-            background-color: #fcfcfb;
-            border: 1px solid #e5e5e5;
-            border-radius: 12px;
-            padding: 20px;
-            height: 80vh;
-            overflow-y: auto;
-            box-shadow: inset 0 0 10px rgba(0,0,0,0.02);
-        }}
-
-        /* Floating Input Bar */
-        .stChatInputContainer {{
-            border: 1px solid #d1d1d1 !important;
-            border-radius: 16px !important;
-            box-shadow: 0 8px 32px rgba(0,0,0,0.06) !important;
-            max-width: 800px !important;
-            margin: 0 auto 20px auto !important;
-        }}
-        
-        header, footer {{ visibility: hidden; }}
-    </style>
-""", unsafe_allow_html=True)
-
-# 2. Model Initialization
+# 1. Load Qwen3 (Cached for efficiency)
 @st.cache_resource
-def load_llm():
-    model_id = "Qwen/Qwen2.5-3B-Instruct" # Highest 2026 performance/efficiency ratio
+def load_qwen3():
+    model_id = "Qwen/Qwen3-1.7B-Instruct"
     tokenizer = AutoTokenizer.from_pretrained(model_id)
-    model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")
-    return model, tokenizer
-
-model, tokenizer = load_llm()
-
-# 3. Sidebar: Context & Settings
-with st.sidebar:
-    st.markdown(f"<h2 style='color: #1a1a1a; font-size: 1.25rem;'>Claude 3.5 Sonnet</h2>", unsafe_allow_html=True)
-    if st.button("+ Start New Chat", use_container_width=True):
+    model = AutoModelForCausalLM.from_pretrained(
+        model_id,
+        torch_dtype="auto",
+        device_map="auto"
+    )
+    return tokenizer, model
+
+tokenizer, model = load_qwen3()
+
+# 2. System Prompt Selection (Main UI, No Sidebar)
+st.title("⚡ Qwen3-1.7B Local Chat")
+
+system_options = {
+    "General Assistant": "You are a helpful and concise assistant.",
+    "Python Expert": "You are an expert Python developer. Provide clean, efficient code with brief explanations.",
+    "Creative Storyteller": "You are a creative writer. Use vivid imagery and engaging narrative styles."
+}
+
+# Horizontal layout for the selector and a clear button
+col1, col2 = st.columns([3, 1])
+with col1:
+    selected_role = st.selectbox("Choose AI Personality:", list(system_options.keys()))
+with col2:
+    if st.button("Clear History", use_container_width=True):
         st.session_state.messages = []
-        st.session_state.artifact_content = ""
         st.rerun()
-    
-    st.divider()
-    st.caption("CAPABILITIES")
-    show_artifacts = st.toggle("Artifacts (Preview)", value=True)
-    st.caption("Recent Artifacts")
-    if st.session_state.get("artifact_content"):
-        st.info("📄 current_code_snippet.py")
-
-# 4. Layout Definition (Chat vs Artifacts)
-if show_artifacts and st.session_state.get("artifact_content"):
-    col_chat, col_art = st.columns([1, 1], gap="large")
-else:
-    col_chat = st.container()
-    col_art = None
-
-# 5. Chat Logic
-if "messages" not in st.session_state: st.session_state.messages = []
-if "artifact_content" not in st.session_state: st.session_state.artifact_content = ""
-
-with col_chat:
-    # Landing View
-    if not st.session_state.messages:
-        st.markdown("<div style='height: 10vh;'></div>", unsafe_allow_html=True)
-        st.markdown("<h1 style='text-align: center; font-weight: 500;'>How can I help you today?</h1>", unsafe_allow_html=True)
 
-    for m in st.session_state.messages:
-        with st.chat_message(m["role"]):
-            st.markdown(m["content"])
+system_prompt = system_options[selected_role]
 
-    if prompt := st.chat_input("Ask Claude..."):
-        st.session_state.messages.append({"role": "user", "content": prompt})
-        with st.chat_message("user"): st.markdown(prompt)
+# 3. Setup Session State
+if "messages" not in st.session_state:
+    st.session_state.messages = []
 
-        with st.chat_message("assistant"):
-            streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
-            msgs = [{"role": "system", "content": "You are Claude. If the user asks for code, provide it clearly."}] + st.session_state.messages[-6:]
-            inputs = tokenizer.apply_chat_template(msgs, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
-            
-            thread = Thread(target=model.generate, kwargs=dict(input_ids=inputs, streamer=streamer, max_new_tokens=1024))
-            thread.start()
-            
-            full_response = st.write_stream(streamer)
-            
-            # Detect if response contains code to trigger "Artifact"
-            if "```" in full_response:
-                code_content = full_response.split("```")[1]
-                st.session_state.artifact_content = code_content
-                st.rerun()
+# Display Chat History
+for message in st.session_state.messages:
+    if message["role"] != "system": # Hide system prompt from UI
+        with st.chat_message(message["role"]):
+            st.markdown(message["content"])
 
-        st.session_state.messages.append({"role": "assistant", "content": full_response})
-
-# 6. Artifact Display (Right Panel)
-if col_art and st.session_state.artifact_content:
-    with col_art:
-        st.markdown("<p style='font-size: 0.8rem; color: #6e6e6e; margin-bottom: 5px;'>ARTIFACT</p>", unsafe_allow_html=True)
-        with st.container(border=True):
-            st.code(st.session_state.artifact_content, language="python")
-            if st.button("Copy to Clipboard"):
-                st.toast("Code copied!")
+# 4. Chat Input & Streaming
+if prompt := st.chat_input("Message Qwen3..."):
+    # Insert system prompt if history is empty
+    if len(st.session_state.messages) == 0:
+        st.session_state.messages.append({"role": "system", "content": system_prompt})
+    
+    # Add User Message
+    st.session_state.messages.append({"role": "user", "content": prompt})
+    with st.chat_message("user"):
+        st.markdown(prompt)
+
+    # Generate Assistant Response
+    with st.chat_message("assistant"):
+        # Prepare input with chat template
+        input_ids = tokenizer.apply_chat_template(
+            st.session_state.messages,
+            tokenize=True,
+            add_generation_prompt=True,
+            return_tensors="pt"
+        ).to(model.device)
+
+        streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
+        
+        generation_kwargs = dict(
+            input_ids=input_ids,
+            streamer=streamer,
+            max_new_tokens=1024,
+            temperature=0.7
+        )
+
+        # Threaded generation for real-time streaming
+        thread = Thread(target=model.generate, kwargs=generation_kwargs)
+        thread.start()
+
+        # Stream the output
+        full_response = st.write_stream(streamer)
+    
+    st.session_state.messages.append({"role": "assistant", "content": full_response})