chat_completion.py

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig


model_name_or_path = "deepseek-ai/DeepSeek-V2-Lite-Chat"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True)
# `device_map` cannot be set to `auto`
model = AutoModelForCausalLM.from_pretrained(
    model_name_or_path,
    trust_remote_code=True,
    torch_dtype=torch.bfloat16).cuda()

model.generation_config = GenerationConfig.from_pretrained(model_name_or_path)
model.generation_config.pad_token_id = model.generation_config.eos_token_id

messages = [
    {"role": "user", "content": "Write a piece of quicksort code in C++"}
]

input_tensor = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
outputs = model.generate(input_tensor.to(model.device), max_new_tokens=100)
result = tokenizer.decode(outputs[0][input_tensor.shape[1]:], skip_special_tokens=True)

print("result", result)