Function run_llm

codes/3_coding_llm.py:179–187

(msg)

def run_llm(msg):
    # vllm
    prompt_token_ids = [tokenizer.apply_chat_template(messages, add_generation_prompt=True) for messages in [msg]]

    outputs = llm.generate(prompt_token_ids=prompt_token_ids, sampling_params=sampling_params)

    completion = [output.outputs[0].text for output in outputs]
    
    return completion[0] 
    

# testing for checking