Is the GGUF file still being uploaded?

#2
by lingyezhixing - opened

The file list is empty, is the repository created by mistake or the file has not been uploaded?

UD Q3_K_XL has just been uploaded one minute ago :)

Hopefully all should be here in the next hour or so because the imatrix takes a while but the quants should be quick.

Unsloth AI org

All done sorry! I was investigating why the perplexity of the model was super high - 180 to 300. It should be OK I think - I verified after using it a bit. Please use:

./llama.cpp/llama-cli -hf unsloth/Hunyuan-A13B-Instruct-GGUF:Q4_K_XL -ngl 99 --jinja --temp 0.7 --top-k 20 --top-p 0.8 --repeat-penalty 1.05

haha it is that high, see https://github.com/ggml-org/llama.cpp/pull/14425, making all confused

Sign up or log in to comment