Skip to content

--main_gpu does not select specified gpu - Intel Arc GPUs #9

@Chadwicksracing

Description

@Chadwicksracing

C:\Users\rtm\Documents>gputopia-worker-opencl-win-64.exe --ln_url silentmeadow15054@getalby.com --force_layers 33 --test_model TheBloke/CodeLlama-7B-Instruct-GGUF:Q4_K_M --debug --main_gpu 0 --tensor_split 0,1,0,1,0,1,0,1,0,1
2023-10-07 10:52:12,482 - ai_worker.main - DEBUG - no nvidia: NVML Shared Library Not Found
{'auth_key': '',
'cl_driver_version': 'OpenCL 3.0 ',
'cl_gpus': [{'clock': 2400,
'clock_unit': 'mhz',
'memory': 8319.0,
'name': 'Intel(R) Arc(TM) A750 Graphics',
'uuid': None},
{'clock': 2400,
'clock_unit': 'mhz',
'memory': 8319.0,
'name': 'Intel(R) Arc(TM) A750 Graphics',
'uuid': None},
{'clock': 2400,
'clock_unit': 'mhz',
'memory': 8319.0,
'name': 'Intel(R) Arc(TM) A750 Graphics',
'uuid': None},
{'clock': 2400,
'clock_unit': 'mhz',
'memory': 8319.0,
'name': 'Intel(R) Arc(TM) A750 Graphics',
'uuid': None},
{'clock': 2400,
'clock_unit': 'mhz',
'memory': 8319.0,
'name': 'Intel(R) Arc(TM) A750 Graphics',
'uuid': None}],
'cpu_count': 16,
'disk_space': 428351,
'ln_address': 'silentmeadow15054@getalby.com',
'ln_url': 'silentmeadow15054@getalby.com',
'nv_driver_version': None,
'nv_gpu_count': None,
'nv_gpus': [],
'vram': 5996646400,
'web_gpus': [],
'worker_id': '',
'worker_version': '0.1.9'}
2023-10-07 10:52:13,051 - ai_worker.main - DEBUG - loading model: TheBloke/CodeLlama-7B-Instruct-GGUF:Q4_K_M
ggml_opencl: selecting platform: 'Intel(R) OpenCL Graphics'
ggml_opencl: selecting device: 'Intel(R) Arc(TM) A750 Graphics'
ggml_opencl: device FP16 support: true
Traceback (most recent call last):
File "ai_worker_main_.py", line 4, in
File "ai_worker\main.py", line 385, in main
File "asyncio\runners.py", line 190, in run
File "asyncio\runners.py", line 118, in run
File "asyncio\base_events.py", line 653, in run_until_complete
File "ai_worker\main.py", line 144, in run
File "ai_worker\main.py", line 114, in test_model
File "ai_worker\main.py", line 201, in load_model
File "llama_cpp\server\app.py", line 342, in create_app
File "llama_cpp\llama.py", line 312, in init
IndexError: invalid index
[12344] Failed to execute script 'main' due to unhandled exception!

image

--tensor_split not working either

Metadata

Metadata

Assignees

No one assigned

    Labels

    bountyWill pay for someone to get this done

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions