build: 4998 (d41f0150) with Apple clang version 16.0.0 (clang-1600.0.26.6) for arm64-apple-darwin24.3.0
llama_model_load_from_file_impl: using device Metal (Apple M3 Pro) - 27647 MiB free
llama_model_loader: loaded meta data with 31 key-value pairs and 339 tensors from ./DeepSeek-R1-Distill-Qwen-7B-IQ3_M.gguf (version GGUF V3 (latest))

Final Winogrande score(750 tasks): 55.6000 +/- 1.8155

llama_perf_context_print:        load time =     232.91 ms
llama_perf_context_print: prompt eval time =   85238.34 ms / 22362 tokens (    3.81 ms per token,   262.35 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 =   86277.57 ms / 22363 tokens
ggml_metal_free: deallocating