cmn  common_param:   - MTL0    : Synthetic GPU (12288 MiB, 12287 MiB free)
common_memory_breakdown_print: |   - MTL0 (Synthetic GPU)    | 12288 = 12287 + (28993 =  4167 +   19641 +    5184) +      -28993 |
common_params_fit_impl: projected to use 28993 MiB of device memory vs. 12287 MiB of free device memory
common_params_fit_impl: cannot meet free memory target of 1024 MiB, need to reduce device memory by 17730 MiB
common_params_fit_impl: context size set by user to 1048576 -> no change
common_fit_params: failed to fit params to free device memory: n_gpu_layers already set by user to 99, abort
load_tensors:  MTL0_Mapped model buffer size =  4167.21 MiB
llama_kv_cache:       MTL0 KV buffer size = 19584.00 MiB
llama_kv_cache:       MTL0 KV buffer size =    57.38 MiB
sched_reserve:       MTL0 compute buffer size =  5184.99 MiB
ggml_metal_synchronize: error: command buffer 0 failed with status 5
ggml_metal_graph_compute: backend is in error state from a previous command buffer failure - recreate the backend to recover
graph_compute: ggml_backend_sched_graph_compute_async failed with error -1
process_ubatch: failed to compute graph, compute status: -1
srv  llama_server: listening
common_memory_breakdown_print: |   - MTL0 (Synthetic GPU)    | 12288 =    0 + (28993 =  4167 +   19641 +    5184) +      -16705 |
