build: 5080 (78a1ba0a) with Apple clang version 16.0.0 (clang-1600.0.26.6) for arm64-apple-darwin24.2.0
llama_model_load_from_file_impl: using device Metal (Apple M3 Pro) - 27647 MiB free
llama_model_loader: loaded meta data with 33 key-value pairs and 292 tensors from ./DeepSeek-R1-Distill-Llama-8B-IQ4_NL.gguf (version GGUF V3 (latest))

Final result: 34.8000 +/- 1.7405
Random chance: 19.8992 +/- 1.4588


llama_perf_context_print:        load time =     293.98 ms
llama_perf_context_print: prompt eval time =  167599.00 ms / 50072 tokens (    3.35 ms per token,   298.76 tokens per second)
llama_perf_context_print:        eval time =       0.00 ms /     1 runs   (    0.00 ms per token,      inf tokens per second)
llama_perf_context_print:       total time =  169473.22 ms / 50073 tokens
ggml_metal_free: deallocating