import gradio as gr
from huggingface_hub import InferenceClient

def respond(
message,
history: list[dict[str, str]],
system_message,
max_tokens,
temperature,
top_p,
hf_token: gr.OAuthToken,
):
"""
Chat adaptado al modelo Nextgengirls/MODELO_BECAS
"""
client = InferenceClient(token=hf_token.token, model="Nextgengirls/MODELO_BECAS")

# El modelo no usa roles tipo "system/user", así que solo concatenamos el texto
prompt = system_message + "\nUsuario: " + message

# Generamos texto con parámetros básicos
response = ""
for message in client.text_generation(
prompt,
max_new_tokens=max_tokens,
stream=True,
temperature=temperature,
top_p=top_p,
):
response += message
yield response


# 🎨 Interfaz igual que antes, solo cambia el modelo detrás
chatbot = gr.ChatInterface(
respond,
type="messages",
additional_inputs=[
gr.Textbox(value="Eres un asistente de becas amable y claro.", label="System message"),
gr.Slider(minimum=1, maximum=512, value=150, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=1.5, value=0.8, step=0.1, label="Temperature"),
gr.Slider(
minimum=0.1,
maximum=1.0,
value=0.9,
step=0.05,
label="Top-p (nucleus sampling)",
),
],
)

with gr.Blocks() as demo:
with gr.Sidebar():
gr.LoginButton()
chatbot.render()

if __name__ == "__main__":
demo.launch()