mirror of
https://github.com/oobabooga/text-generation-webui.git
synced 2024-10-01 01:26:03 -04:00
Use ExLlama_HF for GPTQ models by default
This commit is contained in:
parent
619093483e
commit
09f807af83
@ -110,7 +110,7 @@ def infer_loader(model_name, model_settings):
|
|||||||
if not path_to_model.exists():
|
if not path_to_model.exists():
|
||||||
loader = None
|
loader = None
|
||||||
elif (path_to_model / 'quantize_config.json').exists() or ('wbits' in model_settings and type(model_settings['wbits']) is int and model_settings['wbits'] > 0):
|
elif (path_to_model / 'quantize_config.json').exists() or ('wbits' in model_settings and type(model_settings['wbits']) is int and model_settings['wbits'] > 0):
|
||||||
loader = 'AutoGPTQ'
|
loader = 'ExLlama_HF'
|
||||||
elif (path_to_model / 'quant_config.json').exists() or re.match(r'.*-awq', model_name.lower()):
|
elif (path_to_model / 'quant_config.json').exists() or re.match(r'.*-awq', model_name.lower()):
|
||||||
loader = 'AutoAWQ'
|
loader = 'AutoAWQ'
|
||||||
elif len(list(path_to_model.glob('*.gguf'))) > 0:
|
elif len(list(path_to_model.glob('*.gguf'))) > 0:
|
||||||
|
Loading…
Reference in New Issue
Block a user