Update app.py

app.py

@@ -1,25 +1,40 @@
-import gradio as gr
+from flask import Flask, request, jsonify
 from huggingface_hub import InferenceClient
 import os
 
+app = Flask(__name__)
+
+# Hugging Face Token (set in Space secrets as HF_TOKEN)
 HF_TOKEN = os.getenv("HF_TOKEN")
+client = InferenceClient(model="openai/gpt-oss-20b", token=HF_TOKEN)
+
+@app.route("/chat", methods=["POST"])
+def chat():
+    try:
+        data = request.get_json()
+        user_message = data.get("message", "")
+
+        messages = [
+            {"role": "system", "content": "You are a friendly chatbot."},
+            {"role": "user", "content": user_message}
+        ]
+
+        response = ""
+        for msg in client.chat_completion(
+            messages,
+            max_tokens=256,
+            stream=True,
+            temperature=0.7,
+            top_p=0.95
+        ):
+            if msg.choices and msg.choices[0].delta.content:
+                response += msg.choices[0].delta.content
 
-def respond(message, history):
-    client = InferenceClient(token=HF_TOKEN, model="openai/gpt-oss-20b")
-    messages = [{"role": "system", "content": "You are a friendly chatbot."}]
-    for h in history:
-        messages.append({"role": "user", "content": h[0]})
-        messages.append({"role": "assistant", "content": h[1]})
-    messages.append({"role": "user", "content": message})
+        return jsonify({"reply": response})
 
-    response = ""
-    for msg in client.chat_completion(messages, max_tokens=200, stream=True):
-        if msg.choices and msg.choices[0].delta.content:
-            response += msg.choices[0].delta.content
-    return response
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
 
-# 👇 is tarah karo taake API endpoint expose ho
-demo = gr.ChatInterface(respond)
 
 if __name__ == "__main__":
-    demo.launch()
+    app.run(host="0.0.0.0", port=7860)