app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
# Load your model from Hugging Face Transformers
model_name = "deepseek-ai/DeepSeek-V2-Lite"

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True, torch_dtype=torch.bfloat16).cpu()
model.generation_config = GenerationConfig.from_pretrained(model_name)
model.generation_config.pad_token_id = model.generation_config.eos_token_id
# Define a function to use the model
def math_inference(input_text):
 inputs = tokenizer(input_text, return_tensors="pt")
 output = model.generate(**inputs)
 response = tokenizer.decode(output[0], skip_special_tokens=True)
 return response

# Create a Gradio interface
iface = gr.Interface(
 fn=math_inference,
 inputs=gr.Textbox(prompt="Input math question"),
 outputs=gr.Textbox(prompt="Math answer"),
 layout="vertical",
 title="Math Solver"
)

# Launch the Gradio interface
iface.launch()