|
import gradio as gr |
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("ai-forever/ruGPT-3.5-13B") |
|
model = AutoModelForCausalLM.from_pretrained("ai-forever/ruGPT-3.5-13B") |
|
|
|
|
|
def generate_response(question): |
|
|
|
input_ids = tokenizer.encode(question, return_tensors="pt") |
|
|
|
|
|
output = model.generate(input_ids) |
|
|
|
|
|
decoded_output = tokenizer.decode(output[0], skip_special_tokens=True) |
|
|
|
return decoded_output |
|
|
|
|
|
iface = gr.Interface(fn=generate_response, inputs="text", outputs="text") |
|
|
|
|
|
iface.launch() |
|
|