% llama serve -hf ggml-org/gemma-4-E2B-it-GGUF:Q8_0
0.04.288.612 I cmn common_param: common_params_print_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.04.289.229 W srv llama_server: -----------------
0.04.289.235 W srv llama_server: CORS is set to allow all origins ('*') and no API key is set
0.04.289.236 W srv llama_server: this can be a security risk (cross-origin attacks)
0.04.289.236 W srv llama_server: more info: https://github.com/ggml-org/llama.cpp/pull/25655
0.04.289.237 W srv llama_server: -----------------
Downloading mmproj-gemma-4-E2B-it-Q8_0.gguf ──────────────────────── 100%
Downloading gemma-4-E2B-it-Q8_0.gguf ─────────────────────────────── 100%
7.34.082.011 I srv load_model: loading model 'ggml-org/gemma-4-E2B-it-GGUF:Q8_0'
7.34.113.425 E mtmd_get_memory_usage: error: Failed to load CLIP model from /Users/<user>/.cache/huggingface/hub/models--ggml-org--gemma-4-E2B-it-GGUF/snapshots/858dcdf955fb1b5a43ed2301aea00362fc443a5c/mmproj-gemma-4-E2B-it-Q8_0.gguf
7.34.113.434 E srv load_model: [mtmd] failed to get memory usage of mmproj
7.34.114.075 E llama_model_load_from_file_impl: no backends are loaded. hint: use ggml_backend_load() or ggml_backend_load_all() to load a backend before calling this function
7.34.114.824 E common_fit_params: encountered an error while trying to fit params to free device memory: failed to load model
7.34.114.828 E llama_model_load_from_file_impl: no backends are loaded. hint: use ggml_backend_load() or ggml_backend_load_all() to load a backend before calling this function
7.34.114.829 E cmn common_init_: failed to load model '/Users/<user>/.cache/huggingface/hub/models--ggml-org--gemma-4-E2B-it-GGUF/snapshots/858dcdf955fb1b5a43ed2301aea00362fc443a5c/gemma-4-E2B-it-Q8_0.gguf'
7.34.114.831 E srv load_model: failed to load model, '/Users/<user>/.cache/huggingface/hub/models--ggml-org--gemma-4-E2B-it-GGUF/snapshots/858dcdf955fb1b5a43ed2301aea00362fc443a5c/gemma-4-E2B-it-Q8_0.gguf'
7.34.114.837 I srv operator(): operator(): cleaning up before exit...
7.34.115.083 E srv llama_server: exiting due to model loading error
tested the same with brew and it doesn't have the problem: