ReadyArt/Broken-Tutu-24B-Unslop-v2.0
I took the liberty of actually testing this one before sending it over to you with grand claims this time XD
https://huggingface.co/ReadyArt/Broken-Tutu-24B-Unslop-v2.0
Enjoy!
I think you forgot the grand claims, though? Also, I need to remember the trick of calling it "Broken" - that way, you can never fail.
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#Broken-Tutu-24B-Unslop-v2.0-GGUF for quants to appear.
I took the liberty of actually testing this one before sending it over to you with grand claims this time XD
https://huggingface.co/ReadyArt/Broken-Tutu-24B-Unslop-v2.0
Enjoy!
Wanted to try it, but there are no imatrix quanta ;(
Wanted to try it, but there are no imatrix quanta ;(
Sorry they will come soon. As you can see on https://hf.tst.eu/status.html we are currently going through the backlog 671B models now that we finally fixed imatrix computation of models using MLA. 671B models are so massive that computing the imatrix of them requires all my 3 Servers, 4 GPUs and 896 GiB of RAM so it is unfortunately no possible for us to run any imatrix computation while the imatrix computation of such a model is running. Imatrix quants for this model should appear in around 5 hours. Because of some unexpected technical issues, the timing for this imatrix RPC computation was especially unfortunate. If you don't want to wait, just try the static quants for now. At Q6 there is barely any quality difference.
Wanted to try it, but there are no imatrix quanta ;(
Sorry they will come soon. As you can see on https://hf.tst.eu/status.html we are currently going through the backlog 671B models now that we finally fixed imatrix computation of models using MLA. 671B models are so massive that computing the imatrix of them requires all my 3 Servers, 4 GPUs and 896 GiB of RAM so it is unfortunately no possible for us to run any imatrix computation while the imatrix computation of such a model is running. Imatrix quants for this model should appear in around 5 hours. Because of some unexpected technical issues, the timing for this imatrix RPC computation was especially unfortunate. If you don't want to wait, just try the static quants for now. At Q6 there is barely any quality difference.
Thank you so much for your hard work. It's okay, I'll wait