Update app.py

app.py

@@ -1,70 +1,81 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import chromadb
+from sentence_transformers import SentenceTransformer
+import google.generativeai as genai
+import os
 
+# Get API key from Hugging Face secrets
+GOOGLE_API_KEY = os.environ.get('GOOGLE_API_KEY')
+genai.configure(api_key=GOOGLE_API_KEY)
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
+# Load models (cached)
+print("Loading embedding model...")
+embedding_model = SentenceTransformer('all-MiniLM-L6-v2')
+print("Loading database...")
+chroma_client = chromadb.PersistentClient(path="./deadcells_db_free")
+collection = chroma_client.get_collection(name="deadcells_wiki")
+print("Ready!")
 
-    messages = [{"role": "system", "content": system_message}]
+# Find Gemini model
+available_model = 'gemini-1.5-flash-latest'
 
-    messages.extend(history)
+def get_embedding(text):
+    return embedding_model.encode(text).tolist()
 
-    messages.append({"role": "user", "content": message})
+def chat(message, history):
+    """Handle chat messages"""
+    
+    # Search wiki
+    question_embedding = get_embedding(message)
+    results = collection.query(
+        query_embeddings=[question_embedding],
+        n_results=5
+    )
+    
+    relevant_chunks = results['documents'][0]
+    sources = results['metadatas'][0]
+    
+    if not relevant_chunks:
+        return "I couldn't find any relevant information in the wiki."
+    
+    # Build context
+    context = "\n\n---\n\n".join(relevant_chunks)
+    
+    # Ask Gemini
+    model = genai.GenerativeModel(available_model)
+    prompt = f"""You are a Dead Cells expert. Answer using ONLY the provided wiki content.
 
-    response = ""
+Wiki Content:
+{context}
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+Question: {message}
 
-        response += token
-        yield response
+Answer:"""
+    
+    response = model.generate_content(prompt)
+    answer = response.text
+    
+    # Add sources
+    source_list = "\n\n📚 **Sources:** " + ", ".join([s['source'] for s in sources[:3]])
+    
+    return answer + source_list
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-chatbot = gr.ChatInterface(
-    respond,
-    type="messages",
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+# Create Gradio interface
+demo = gr.ChatInterface(
+    fn=chat,
+    title="🎮 Dead Cells Wiki Bot",
+    description="Ask me anything about Dead Cells! I answer using the official wiki.",
+    examples=[
+        "What achievements are there for beating bosses?",
+        "Tell me about the Hand of the King",
+        "How does malaise work?",
+        "What are boss stem cells?",
     ],
+    theme="soft",
+    retry_btn=None,
+    undo_btn=None,
+    clear_btn="Clear Chat"
 )
 
-with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
-    chatbot.render()
-
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()