import os
os.environ["CUDA_VISIBLE_DEVICES"] = "3"
from vllm import LLM, SamplingParams

# Sample prompts.
prompts = [
    "You are a helpful assistant.\n\nUser:张亮的爸爸叫张明，张明的爸爸有三个孩子，大儿子叫张大，二儿子叫张昊，三儿子叫什么？\nAssistant:",
    "You are a helpful assistant.\n\nUser:你好\nAssistant:",
    "You are a helpful assistant.\n\nUser:1+1等于几\nAssistant:",
    "You are a helpful assistant.\n\nUser:你是谁\nAssistant:",
]
# Create a sampling params object.
sampling_params = SamplingParams(temperature=0, top_p=1, presence_penalty=0.9, max_tokens=1024)

# Create an LLM.
llm = LLM(model="/home/majiahui/project/models-llm/openbuddy-mistral-7b-v13.1", trust_remote_code=True)
# Generate texts from the prompts. The output is a list of RequestOutput objects
# that contain the prompt, generated text, and other information.
outputs = llm.generate(prompts, sampling_params)
# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")