Iván Baldo
Iván Baldo
Maybe skip QTIP and go straight to https://github.com/Cornell-RelaxML/yaqa-quantization ? Since these are GPL-V3 I guess they need to be implemented from the papers and not from the code....
If it's fixed then maybe this issue can be closed? @AttaInsularis can you confirm?
Having the same issue on bigbench_generate_until with v0.4.8 :-(
Related to this, the users can't download the source document, see https://github.com/open-webui/open-webui/discussions/7578 .