Update app.py

app.py

@@ -1,38 +1,17 @@
-import gradio as gr
+from fastapi import FastAPI
+from pydantic import BaseModel
 from transformers import pipeline
 
-# ===========================
-# LOAD MODEL
-# ===========================
-model_id = "HuggingFaceH4/zephyr-7b-beta"
-ai = pipeline("text-generation", model=model_id, max_new_tokens=200)
+ai = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-beta", max_new_tokens=200)
+app = FastAPI()
 
-# ===========================
-# API FUNCTION
-# ===========================
-def chat_api(message):
+class Msg(BaseModel):
+    message: str
+
+@app.post("/chat")
+def chat(msg: Msg):
     try:
-        output = ai(message)[0]["generated_text"]
+        output = ai(msg.message)[0]["generated_text"]
         return {"reply": output}
     except:
-        return {"reply": "ERROR"}
-
-# ===========================
-# GRADIO APP (UI + API)
-# ===========================
-with gr.Blocks() as demo:
-
-    gr.Markdown("# 🚀 Space 1 — Ghost Model")
-
-    inp = gr.Textbox(label="Message")
-    out = gr.Textbox(label="Reply")
-    btn = gr.Button("Send")
-    btn.click(lambda x: chat_api(x)["reply"], inp, out)
-
-    # API endpoint
-    gr.Interface(fn=chat_api, inputs=gr.Textbox(), outputs=gr.JSON(), live=False, title="API Endpoint", description="Use /chat JSON POST")
-
-# ===========================
-# LAUNCH
-# ===========================
-demo.launch(server_name="0.0.0.0", server_port=7860)
+        return {"reply": "ERROR"}