failed to load Core ML model, Timeout Error

[GrJona]

What happened?

MacOS 2019 intel chip,
Doesn't work both on default model and ivrit model.

Steps to reproduce

Install x64 intel version, try to transcribe anything

What OS are you seeing the problem on?

MacOS

Relevant log output

options: {
  "path": "/Users/***/Documents/2025-01-24 20-15-40.wav",
  "lang": "he",
  "verbose": false,
  "n_threads": 4,
  "init_prompt": "",
  "temperature": 0.4,
  "translate": null,
  "max_text_ctx": null,
  "word_timestamps": true,
  "max_sentence_len": 32
}

Caused by:
   0: failed to transcribe
   1: Generic whisper error. Varies depending on the function. Error code: -6

Location:
    core/src/transcribe.rs:303:38
App Version: vibe 3.0.1
Commit Hash: 67820a0
Arch: x86_64
Platform: macos
Kernel Version: 14.4.0
OS: macos
OS Version: 14.4.0
Cuda Version: n/a
Models: ggml-medium-q8_0.bin, ggml-ivrit-v2-d4.bin
Default Model: ggml-ivrit-v2-d4.bin"
Cargo features: 


CPU feature detection is not supported on this architecture.


<details>
<summary>logs</summary>


whisper_init_state: failed to load Core ML model from '/Users/**/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
cmd: "/Applications/vibe.app/Contents/MacOS/../Resources/ffmpeg" "-i" "/var/folders/vl/5qz7kg4s3dg3_v5w1x_4tmxr0000gn/T/vibe_temp_2025-01-24/ZHKQ0wOvlE.wav" "-ar" "16000" "-ac" "1" "-c:a" "pcm_s16le" "/var/folders/vl/5qz7kg4s3dg3_v5w1x_4tmxr0000gn/T/vibe_temp_2025-01-24/2025-01-24 20-15-40.wav" "-hide_banner" "-y" "-loglevel" "error"
whisper_init_state: failed to load Core ML model from '/Users/**/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
whisper_init_state: failed to load Core ML model from '/Users/**/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode

</details>

[thewh1teagle]

What if you place the file there in models folder?

Download it from hugging face
ggml-medium-encoder.mlmodelc
Open vibe settings > download models > at the bottom 'want more' > search for the file and download> open folder from settings and place there and transcribe again

[GrJona]

@thewh1teagle seems to fix the issue, thanks