LlamaIndex
Log in

Find answers from the community

Updated 2 months ago

Why does LlamaIndex make requests to

ddoughboy
·
Why does LlamaIndex make requests to huggingface.co for config files every single time I fire it up? How can I disable this?
Plain Text
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): huggingface.co:443
Starting new HTTPS connection (1): huggingface.co:443
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/config.json HTTP/1.1" 200 0
https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/config.json HTTP/1.1" 200 0
INFO:datasets:PyTorch version 2.1.2 available.
PyTorch version 2.1.2 available.
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/generation_config.json HTTP/1.1" 200 0
https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/generation_config.json HTTP/1.1" 200 0
DEBUG:urllib3.connectionpool:https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/tokenizer_config.json HTTP/1.1" 200 0
https://huggingface.co:443 "HEAD /TheBloke/WestSeverus-7B-DPO-AWQ/resolve/main/tokenizer_config.json HTTP/1.1" 200 0
init took 4.84630s
L
d
3 comments
LLogan M
Huggingface likes to check for updated model weights/configs

If you actually download the model, and point it towards the downloaded path, it won't check.

There might be some other huggingface-specific workaround I'm not aware of
ddoughboy
Ah, okay. Thank you. I just quantized mlabonne/Daredevil-7B with AWQ, and I'm no longer seeing the HTTP requests—probably because I saved it within my project and not within ~/.cache/huggingface/hub?
ddoughboy
Anyway, I am still seeing this in my terminal, but I already have PyTorch 2.1.2 installed. Weird.
Plain Text
INFO:datasets:PyTorch version 2.1.2 available.
PyTorch version 2.1.2 available.
Add a reply
Sign up and join the conversation on Discord