search

city96 / ComfyUI-GGUF

GGUF Quantization support for native ComfyUI models
Apache License 2.0
796 stars 45 forks source link

convert.py #119

Open yggdrasil75 opened 2 hours ago

yggdrasil75 commented 2 hours ago

Traceback (most recent call last): File "/gits/ComfyUI/custom_nodes/ComfyUI-GGUF/tools/convert.py", line 165, in handle_tensors data = gguf.quants.quantize(data, data_qtype) AttributeError: module 'gguf.quants' has no attribute 'quantize'

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/gits/ComfyUI/custom_nodes/ComfyUI-GGUF/tools/convert.py", line 195, in handle_tensors(path, writer, state_dict) File "/gits/ComfyUI/custom_nodes/ComfyUI-GGUF/tools/convert.py", line 166, in handle_tensors except (AttributeError, gguf.QuantError) as e: AttributeError: module 'gguf' has no attribute 'QuantError'

get this error when trying to convert a model to gguf. could more info about what is needed to convert models be added?

yggdrasil75 commented 2 hours ago

oh. the requirements is wrong. needs 0.10, 0.9.1 is missing a few functions used by the script it appears.