Update app.py

app.py

@@ -2,14 +2,17 @@ from fastapi import FastAPI, Request
 from fastapi.responses import JSONResponse
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
+import gradio as gr
+import requests
+import threading
 
 app = FastAPI()
 
-# Load model and tokenizer
+# Load model and tokenizer once
 tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-medium")
 model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-medium")
 
-# In-memory history per user
+# In-memory chat history by user
 chat_history = {}
 
 @app.get("/")
@@ -35,7 +38,31 @@ async def chat(request: Request):
     chat_history[user_id] = [bot_input_ids, output_ids]
     return JSONResponse({"reply": response})
 
-# Only needed if running locally, not in Hugging Face Space
-if __name__ == "__main__":
-    import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=7860)
+# Gradio UI to call your /ai endpoint easily via browser
+def gradio_chat(user_input, user_id="default"):
+    if not user_input:
+        return "Please enter some text."
+    url = f"https://Trigger82--API.hf.space/ai?query={user_input}&user_id={user_id}"
+    try:
+        res = requests.get(url)
+        if res.status_code == 200:
+            return res.json().get("reply", "No reply")
+        return f"Error: {res.status_code}"
+    except Exception as e:
+        return f"Exception: {e}"
+
+iface = gr.Interface(
+    fn=gradio_chat,
+    inputs=[gr.Textbox(label="Your Message"), gr.Textbox(label="User ID", value="default")],
+    outputs="text",
+    title="Chat with DialoGPT API",
+    description="Type your message and user id to chat with the model."
+)
+
+# Launch Gradio app in a thread alongside FastAPI
+def run_gradio():
+    iface.launch(server_name="0.0.0.0", server_port=7861, share=False)
+
+threading.Thread(target=run_gradio).start()
+
+# No need for uvicorn.run here on Spaces; it manages startup automatically