https://huggingface.co/PrunaAI/PJMixers-Dev-Qwen2.5-RomboTiesTest-7B-bnb-8bit-smashed
#791
by
jackangel
- opened
Apparently it is the top 1 on 7B - but it is already gone from PMMixers-Dev space - https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?params=3%2C7
The model you linked is already quantized and so will almost certainly fail the GGUF conversion but I queued it anyways.
As expected, it failed. The original model from https://huggingface.co/PJMixers-Dev/Qwen2.5-RomboTiesTest-7B was taken down by the author so we can't quantize it. The only way for us to quantize this model is for the original author to reupload his original unquantized model. You could contact the original author which probably is xzuyn and ask him to do so and if he still has it somewhere he might be able to help.
Thank you.
jackangel
changed discussion status to
closed