bug: Doesn't load local models using unicode paths?

[r57zone]

• I have searched the existing issues

Current behavior

If try to import the file "D:\Софт\Windows\Базы данных\LM Studio - AI чаты общения\Модели\lmstudio-community\Meta-Llama-3.1-8B-Instruct-GGUF\Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf" it is imported, but the model does not run.

If import "D:\Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf" it works fine.

Minimum reproduction step

1. Settings -> Import model
2. Drag and drop "D:\Софт\Windows\Базы данных\LM Studio - AI чаты общения\Модели\lmstudio-community\Meta-Llama-3.1-8B-Instruct-GGUF\Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf".
3. Start model in list.

Expected behavior

Work on any paths

Screenshots / Logs

2024-07-25T09:42:15.846Z [CORTEX]::Debug: Killing PID 13700
2024-07-25T09:42:15.846Z [CORTEX]::Debug: Request to kill cortex
2024-07-25T09:42:15.982Z [CORTEX]::Debug: cortex process is terminated
2024-07-25T09:42:16.054Z [CORTEX]::CPU information - 7
2024-07-25T09:42:16.055Z [CORTEX]::Debug: Killing PID 13700
2024-07-25T09:42:16.054Z [CORTEX]::Debug: Request to kill cortex
2024-07-25T09:42:16.159Z [CORTEX]::Failed to kill PID - sending request to kill
2024-07-25T09:42:16.161Z [CORTEX]::Debug: cortex process is terminated
2024-07-25T09:42:16.161Z [CORTEX]::Debug: Spawning cortex subprocess...
2024-07-25T09:42:16.161Z [APP]::C:\Users\r57zone\jan\extensions\@janhq\inference-cortex-extension\dist\bin\win-cpu
2024-07-25T09:42:16.161Z [CORTEX]::Debug: Spawn cortex at path: C:\Users\r57zone\jan\extensions\@janhq\inference-cortex-extension\dist\bin\win-cpu\cortex-cpp.exe, and args: 1,127.0.0.1,3928
2024-07-25T09:42:16.261Z [CORTEX]::Debug: cortex exited with code: 1
2024-07-25T09:42:16.265Z [CORTEX]::Debug: Loading model with params {"cpu_threads":7,"ctx_len":2048,"embedding":false,"prompt_template":"{system_message}\n### Instruction: {prompt}\n### Response:","llama_model_path":"D:\\Софт\\Windows\\Базы данных\\LM Studio - AI чаты общения\\Модели\\lmstudio-community\\Meta-Llama-3.1-8B-Instruct-GGUF\\Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf","system_prompt":"","user_prompt":"\n### Instruction: ","ai_prompt":"\n### Response:","model":"Meta-Llama-3.1-8B-Instruct-Q4_K_M-1","ngl":100}
2024-07-25T09:42:16.265Z [CORTEX]::Debug: cortex is ready
2024-07-25T09:42:16.268Z [CORTEX]::Debug: 20240725 09:42:16.181000 UTC 924 INFO  cortex-cpp version: default_version - main.cc:73
20240725 09:42:16.181000 UTC 924 INFO  cortex.llamacpp version: 0.1.20-30.06.24 - main.cc:78
20240725 09:42:16.181000 UTC 924 INFO  Server started, listening at: 127.0.0.1:3928 - main.cc:81
20240725 09:42:16.181000 UTC 924 INFO  Please load your model - main.cc:82
20240725 09:42:16.181000 UTC 924 INFO  Number of thread is:12 - main.cc:89
20240725 09:42:16.265000 UTC 11376 INFO  CPU instruction set: fpu = 1| mmx = 1| sse = 1| sse2 = 1| sse3 = 1| ssse3 = 1| sse4_1 = 1| sse4_2 = 1| pclmulqdq = 1| avx = 1| avx2 = 1| avx512_f = 0| avx512_dq = 0| avx512_ifma = 0| avx512_pf = 0| avx512_er = 0| avx512_cd = 0| avx512_bw = 0| has_avx512_vl = 0| has_avx512_vbmi = 0| has_avx512_vbmi2 = 0| avx512_vnni = 0| avx512_bitalg = 0| avx512_vpopcntdq = 0| avx512_4vnniw = 0| avx512_4fmaps = 0| avx512_vp2intersect = 0| aes = 1| f16c = 1| - server.cc:277
20240725 09:42:16.265000 UTC 11376 INFO  Loaded engine: cortex.llamacpp - server.cc:305
20240725 09:42:16.265000 UTC 11376 INFO  cortex.llamacpp version: default_version - llama_engine.cc:159
20240725 09:42:16.265000 UTC 11376 ERROR Could not find model in path D:\Софт\Windows\Базы данных\LM Studio - AI чаты общения\Модели\lmstudio-community\Meta-Llama-3.1-8B-Instruct-GGUF\Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf - llama_engine.cc:332

2024-07-25T09:42:16.270Z [CORTEX]::Debug: 20240725 09:42:16.265000 UTC 11376 DEBUG [LoadModelImpl] cache_type: f16 - llama_engine.cc:356
20240725 09:42:16.265000 UTC 11376 DEBUG [LoadModelImpl] Enabled Flash Attention - llama_engine.cc:365
20240725 09:42:16.265000 UTC 11376 DEBUG [LoadModelImpl] stop: null
 - llama_engine.cc:386
{"timestamp":1721900536,"level":"INFO","function":"LoadModelImpl","line":409,"message":"system info","n_threads":7,"total_threads":12,"system_info":"AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 0 | "}

2024-07-25T09:42:16.271Z [CORTEX]::Debug: Load model success with response {}
2024-07-25T09:42:16.271Z [CORTEX]::Debug: {"timestamp":1721900536,"level":"ERROR","function":"LoadModel","line":183,"message":"llama.cpp unable to load model","model":""}
20240725 09:42:16.265000 UTC 11376 ERROR Error loading the model - llama_engine.cc:414

2024-07-25T09:42:16.271Z [CORTEX]::Error: llama_model_load: error loading model: llama_model_loader: failed to load model from 

llama_load_model_from_file: failed to load model
llama_init_from_gpt_params: error: failed to load model ''

2024-07-25T09:42:16.272Z [CORTEX]::Debug: Validate model state with response 409
2024-07-25T09:42:16.272Z [CORTEX]::Debug: Validate model state failed with response "Conflict"
2024-07-25T09:42:16.272Z [CORTEX]::Error: Validate model status failed

Jan version

0.5.2

In which operating systems have you tested?

• macOS
• Windows
• Linux

Environment details

No response

[r57zone]

If there are Russian letters in the folder names, then the model does not start

[Van-QA]

@r57zone,
Sorry for the inconvenience,
The issue is related to #1713, dev team did spend time investigating for a solution but we are working on a refactoring version of the Jan x Cortex, so it might take a while for the issue to be resolved.
Feel free to follow up via the related issue.