app.py

import gradio as gr

def conversation(prompt="", max_tokens=128):
  # Integrate your Bloom 3b model here to generate response based on prompt and max_tokens
  # Replace this with the actual call to your Bloom 3b model
  gr.load("models/CreitinGameplays/bloom-3b-conversational")
  response = "Bloom 3b is currently unavailable. Try again later!"
  return response

interface = gr.Interface(
  fn=conversation,
  inputs=[
      gr.Textbox(label="Text Prompt", value="<|system|> You are a helpful AI assistant </s> <|prompter|> What is an AI? </s> <|assistant|>"),
      gr.Slider(minimum=1, maximum=1024, label="Max New Tokens", value=128),
  ],
  # No outputs are specified as the response will be printed in the console.
  title="Bloom 3b Conversational Assistant",
  description="Talk to Bloom 3b using a text prompt and adjust the maximum number of tokens for response generation.",
)

interface.launch()