Update app.py

app.py

@@ -1,75 +1,32 @@
-import os
-import gradio as gr
+from fastapi import FastAPI
+from pydantic import BaseModel
 from llama_cpp import Llama
-from huggingface_hub import hf_hub_download
 
-# =========================
-# LOAD MODEL
-# =========================
-model_path = hf_hub_download(
-    repo_id="Mikecode123/ALX",
-    filename="qwen2-1_5b-instruct-q4_0.gguf",
-    token=os.getenv("HF_TOKEN")
-)
+app = FastAPI()
 
+# Load model once on startup
 llm = Llama(
-    model_path=model_path,
-    n_ctx=512,
-    n_threads=1
+    model_path="qwen2-1_5b-instruct-q4_0.gguf",
+    n_ctx=2048,
+    n_threads=2
 )
 
-# =========================
-# CHAT FUNCTION (STRICT MESSAGES FORMAT)
-# =========================
-def chat(message, history):
-    history = history or []
-
-    messages = []
-
-    # SAFE: ensure correct format
-    for msg in history:
-        if isinstance(msg, dict):
-            messages.append({
-                "role": msg.get("role", ""),
-                "content": str(msg.get("content", ""))
-            })
+# Request body structure
+class ChatRequest(BaseModel):
+    message: str
 
-    # add user message
-    messages.append({
-        "role": "user",
-        "content": str(message)
-    })
+@app.get("/")
+def home():
+    return {"status": "AI server running"}
 
-    # inference
-    output = llm.create_chat_completion(
-        messages=messages,
+@app.post("/chat")
+def chat(req: ChatRequest):
+    output = llm.create_completion(
+        prompt=req.message,
         max_tokens=300,
         temperature=0.7
     )
 
-    response = output["choices"][0]["message"]["content"]
-
-    # append assistant reply
-    messages.append({
-        "role": "assistant",
-        "content": response
-    })
-
-    return "", messages
-
-# =========================
-# UI (GRADIO 4 SAFE STYLE BUT BACKWARD COMPATIBLE)
-# =========================
-with gr.Blocks() as demo:
-    gr.Markdown("# 🧠 Living Legend AI Chatbot")
-
-    chatbot = gr.Chatbot()  # IMPORTANT FIX
-    msg = gr.Textbox()
-    clear = gr.Button("Clear")
-
-    msg.submit(chat, [msg, chatbot], [msg, chatbot])
-
-    clear.click(lambda: [], None, chatbot)
-
-demo.queue()
-demo.launch(server_name="0.0.0.0", server_port=7860)
+    return {
+        "response": output["choices"][0]["text"].strip()
+    }