Request to Delete and Reupload GGUF Quantized Models
Dear mradermacher,
I would like to request the deletion of the following GGUF quantized models, which I believe are completely damaged:
- Lunzima/NQLSG-Qwen2.5-14B-Base
- Lunzima/NQLSG-Qwen2.5-14B-Chatbot
- Lunzima/NQLSG-Qwen2.5-14B-Coder
Additionally, I have identified that the model below, while previously damaged, has been repaired. Could you please reupload the GGUF quantized version?
- Lunzima/NQLSG-Qwen2.5-14B-MegaFusion
Thank you for your assistance.
Best regards,
Lun Zima
Done as requested. I nuked all the above mentioned repositories (including the one you wanted to requant) and forcefully requed Lunzima/NQLSG-Qwen2.5-14B-MegaFusion.
You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#NQLSG-Qwen2.5-14B-MegaFusion-GGUF for quants to appear.
@mradermacher Here the commands I used in case I did anything wrong but everything looks good to me.
nico1 ~# llmc nukerepo NQLSG-Qwen2.5-14B-Base- NQLSG-Qwen2.5-14B-Chatbot- NQLSG-Qwen2.5-14B-Coder- NQLSG-Qwen2.5-14B-MegaFusion-
NQLSG-Qwen2.5-14B-Base-GGUF...
NQLSG-Qwen2.5-14B-Base-i1-GGUF... RepositoryNotFoundError
NQLSG-Qwen2.5-14B-Chatbot-GGUF...
NQLSG-Qwen2.5-14B-Chatbot-i1-GGUF... RepositoryNotFoundError
NQLSG-Qwen2.5-14B-Coder-GGUF...
NQLSG-Qwen2.5-14B-Coder-i1-GGUF... RepositoryNotFoundError
NQLSG-Qwen2.5-14B-MegaFusion-GGUF...
NQLSG-Qwen2.5-14B-MegaFusion-i1-GGUF... RepositoryNotFoundError
nico1 ~# llmc add force -2000 si https://huggingface.co/Lunzima/NQLSG-Qwen2.5-14B-MegaFusion
submit tokens: ["force","-2000","static","imatrix","https://huggingface.co/Lunzima/NQLSG-Qwen2.5-14B-MegaFusion"]
https://huggingface.co/Lunzima/NQLSG-Qwen2.5-14B-MegaFusion
["https://huggingface.co/Lunzima/NQLSG-Qwen2.5-14B-MegaFusion",["s","0"],1739401723],
https://huggingface.co/Lunzima/NQLSG-Qwen2.5-14B-MegaFusion already in llmjob.submit.txt
forcing.
like a pro :) only one caveat: nukerepo only nukes the repositories, not the cached imatrix file. there were no imatrix quants of these, so all should be perfect. nukeall does (try) to delete the imatrix file, should that be necessary.
i mention that because i fell for this a few times.