0.02.934.812 I load_tensors: offloaded 0/6 layers to GPU
0.02.934.812 I load_tensors:   CPU_Mapped model buffer size =     1.12 MiB
0.02.934.861 I llama_context: n_ctx                 = 512
0.02.934.861 I llama_context: n_ctx_seq             = 512
0.02.934.862 I llama_context: flash_attn            = auto
0.02.934.917 I llama_context:        CPU  output buffer size =     0.00 MiB
0.02.934.938 I llama_kv_cache:        CPU KV buffer size =     0.31 MiB
0.02.934.953 I llama_kv_cache: size =    0.31 MiB (   512 cells,   5 layers,  1/1 seqs), K (f16):    0.16 MiB, V (f16):    0.16 MiB
0.02.935.388 I sched_reserve:        CPU compute buffer size =     2.02 MiB
