1from llama_cpp import Llama
2llm = Llama(
3 model_path="meta-llama-3-8b-instruct.Q8_0.gguf",
4 # n_gpu_layers=-1, # Uncomment to use GPU acceleration
5 # seed=1337, # Uncomment to set a specific seed
6 # n_ctx=2048, # Uncomment to increase the context window
7)
8output = llm(
9 "Q: Name the planets in the solar system? A: ", # Prompt
10 max_tokens=50, # Generate up to 50 tokens, set to None to generate up to the end of the context window
11 stop=["Q:", "\n"], # Stop generating just before the model would generate a new question
12 echo=True # Echo the prompt back in the output
13)
14output
Llama.generate: prefix-match hit
llama_print_timings: load time = 7770.49 ms
llama_print_timings: sample time = 100.16 ms / 40 runs ( 2.50 ms per token, 399.35 tokens per second)
llama_print_timings: prompt eval time = 0.00 ms / 1 tokens ( 0.00 ms per token, inf tokens per second)
llama_print_timings: eval time = 35214.73 ms / 40 runs ( 880.37 ms per token, 1.14 tokens per second)
llama_print_timings: total time = 35895.91 ms / 41 tokens
{'id': 'cmpl-01e2feb3-c0ff-4a6e-8ca4-b8bf2172da01',
'object': 'text_completion',
'created': 1713912080,
'model': 'meta-llama-3-8b-instruct.Q8_0.gguf',
'choices': [{'text': 'Q: Name the planets in the solar system? A: 1. Mercury, 2. Venus, 3. Earth, 4. Mars, 5. Jupiter, 6. Saturn, 7. Uranus, 8. Neptune.',
'index': 0,
'logprobs': None,
'finish_reason': 'stop'}],
'usage': {'prompt_tokens': 13, 'completion_tokens': 40, 'total_tokens': 53}}
