import os
from transformers import AutoModelForCausalLM, AutoTokenizer

# Setze das Cache-Verzeichnis
os.environ['TRANSFORMERS_CACHE'] = 'cache'

model_name = "Monero/WizardLM-Uncensored-SuperCOT-StoryTelling-30b"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Funktion zur Textgenerierung definieren
def generate_text(prompt):
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(inputs["input_ids"], max_length=100)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

import gradio as gr

# Gradio-Interface erstellen
iface = gr.Interface(
    fn=generate_text,
    inputs="text",
    outputs="text",
    title="WizardLM Uncensored SuperCOT StoryTelling"
)

# Interface mit öffentlichem Link starten
iface.launch(share=True)