Skip to content

BUG: Function kernel_mul_mv_ext_bf16_f32_r1_2 was not found in the library #173

Description

@codingl2k1
0.00.001.219 I common_init_result: (for bugs during this step try to reproduce them with -fit off, or provide --verbose logs if the bug only occurs with -fit on)
2026-07-27T13:10:48.905Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58067 - "GET /v1/models/gemma-4/progress HTTP/1.1" 200
2026-07-27T13:10:48.906Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58066 - "GET /v1/models/gemma-4/replicas HTTP/1.1" 200
0.00.490.421 W load: control-looking token:     50 '<|tool_response>' was not control-type; this is probably a bug in the model. its type will be overridden
0.00.492.586 W load: control-looking token:    212 '</s>' was not control-type; this is probably a bug in the model. its type will be overridden
0.00.495.427 W load: control-looking token:      1 '<eos>' was not control-type; this is probably a bug in the model. its type will be overridden
0.00.498.273 W load: special_eog_ids contains '<|tool_response>', removing '</s>' token from EOG list
2026-07-27T13:10:49.908Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58066 - "GET /v1/models/gemma-4/progress HTTP/1.1" 200
2026-07-27T13:10:49.909Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58067 - "GET /v1/models/gemma-4/replicas HTTP/1.1" 200
0.01.517.539 W llama_context: n_ctx_seq (32768) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
2026-07-27T13:10:50.906Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58067 - "GET /v1/models/gemma-4/progress HTTP/1.1" 200
2026-07-27T13:10:50.907Z INFO uvicorn.access pid:73791 role:local address:0.0.0.0:9997 node:MacBook-Pro-3.local 127.0.0.1:58066 - "GET /v1/models/gemma-4/replicas HTTP/1.1" 200
0.02.265.762 I common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
0.02.275.073 E ggml_metal_library_compile_pipeline: failed to compile pipeline: base = 'kernel_mul_mv_ext_bf16_f32_r1_2', name = 'kernel_mul_mv_ext_bf16_f32_r1_2_nsg=2_nxpsg=16_ne12=1_r2=1_r3=1'
0.02.275.115 E ggml_metal_library_compile_pipeline: Error Domain=MTLLibraryErrorDomain Code=5 "Function kernel_mul_mv_ext_bf16_f32_r1_2 was not found in the library" UserInfo={NSLocalizedDescription=Function kernel_mul_mv_ext_bf16_f32_r1_2 was not found in the library}
2026-07-27T13:10:50.987Z ERROR xinference.core.worker pid:73843 role:local address:0.0.0.0:52188 node:MacBook-Pro-3.local Failed to load model gemma-4-rep0

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions