Spaces:
Sleeping
Sleeping
File size: 794 Bytes
11f8d70 decedce af71772 11f8d70 decedce af71772 decedce af71772 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 |
import gradio as gr
import spaces
import subprocess
# from transformers import AutoTokenizer, AutoModelForCausalLM
result = subprocess.run(["python3", "-m", "pip", "install", "transformers==4.34.0"], shell=True, capture_output=True, text=True)
print(result.stdout)
model = None
def greet(name):
return "Hello " + name + "!!"
@spaces.GPU
def load_model():
print(f"Loading model...")
model_path = "meta-llama/Meta-Llama-3-8B"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.bfloat16,
device_map="auto",
)
def main():
load_model()
demo = gr.Interface(fn=greet, inputs="text", outputs="text")
demo.launch()
if __name__ == "__main__":
main()
|