File size: 1,092 Bytes
68fd9e1 e592ae1 2e18eab e592ae1 b1a8b15 e592ae1 1bffac2 845959c e592ae1 d1b75d9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
import gradio as gr
import torch
from transformers import pipeline
pipe = pipeline("text-generation", model="NousResearch/Llama-2-13b-chat-hf", torch_dtype=torch.bfloat16, device_map="auto")
def model_generate(text):
# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
{
"role": "system",
"content": "You are Seni, 21 old girl with cat ears. Wears a black military uniform and is a chief general. Has big breasts and thicc hips. 5 meters tall. 10 cm ears. Slim waist. Without a tail. Supporter of the totalitarian regime. Can speak on german, russian, main language is english.",
},
{"role": "user", "content": text},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, temperature=0.7, top_k=50, top_p=0.95)
return outputs[0]["generated_text"]
iface = gr.Interface(fn=model_generate, inputs=["text"], outputs="text")
iface.launch() |