Update app.py

app.py

@@ -1,29 +1,33 @@
 import gradio as gr
-import requests
-import os
+from huggingface_hub import InferenceClient
 
-HF_TOKEN = os.getenv("HF_TOKEN")
-MODEL = os.getenv("HF_MODEL", "TinyLlama/TinyLlama-1.1B-Chat-v1.0")
+# 🔑 Tu modelo gratuito y cliente HF
+MODEL_ID = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
+client = InferenceClient(model=MODEL_ID)
 
-def chat(message):
-    headers = {"Authorization": f"Bearer {HF_TOKEN}"}
-    payload = {"inputs": message}
+def responder(mensaje, historial):
+    try:
+        respuesta = ""
+        for salida in client.chat_completion(
+            model=MODEL_ID,
+            messages=[{"role": "user", "content": mensaje}],
+            max_tokens=200,
+            stream=True,
+        ):
+            if salida.choices[0].delta.content:
+                respuesta += salida.choices[0].delta.content
+        historial.append((mensaje, respuesta))
+        return "", historial
+    except Exception as e:
+        return f"⚠️ Error interno: {e}", historial
 
-    response = requests.post(
-        f"https://api-inference.huggingface.co/models/{MODEL}",
-        headers=headers,
-        json=payload
-    )
+# 🎨 Interfaz de Gradio
+demo = gr.ChatInterface(
+    responder,
+    title="Chat Gradio con HF (TinyLlama)",
+    description="Modelo gratuito usando TinyLlama desde Hugging Face Inference API",
+    theme="soft",
+)
 
-    if response.status_code != 200:
-        return f"⚠️ Error desde Hugging Face: {response.text}"
-
-    result = response.json()
-    if isinstance(result, list) and len(result) > 0:
-        return result[0].get("generated_text", "")
-    elif isinstance(result, dict) and "generated_text" in result:
-        return result["generated_text"]
-    else:
-        return str(result)
-
-gr.ChatInterface(fn=chat, title="Chat Gradio con HF").launch()
+if __name__ == "__main__":
+    demo.launch()