HauhauCS

/

Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive

like 74

It looks like the manifest files for some variants (e.g. Q4_K_P) are missing, therefore they are not shown on the main model page and also llama.cpp fails to retrieve them. Can you please check?

What manifest files do you mean exactly? Everything is up and working properly. Excluding 1 HuggingFace bug that won't list all quants in the widget directly and makes you have to go here to see them all: https://huggingface.co/HauhauCS/Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive/tree/main

error from HF API (https://huggingface.co/v2/HauhauCS/Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive/manifests/Q4_K_P), response code: 400, data: {"error":"The specified tag is not a valid quantization scheme. Please use another tag or "latest""}

This is the error I'm getting. I assumed that the manifest json file for Q4_K_P variant is not present. Apologies if I misunderstood it.

error from HF API (https://huggingface.co/v2/HauhauCS/Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive/manifests/Q4_K_P), response code: 400, data: {"error":"The specified tag is not a valid quantization scheme. Please use another tag or "latest""}

This is the error I'm getting. I assumed that the manifest json file for Q4_K_P variant is not present. Apologies if I misunderstood it.

Ahh, yeah that makes sense.HuggingFace's API doesn't recognize K_P as a valid quantization tag yet. It's a custom quant type. You'll need to download directly using the filename instead of the tag.

Try: huggingface-cli download HauhauCS/Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive Qwen3.5-122B-A10B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf --local-dir .

I've already opened up a support ticket/request for this but it might be an idea for others to do so as well to show HF there is interest :)

Thank you for the explanation and for your work! Will download the gguf directly. Already tried Q4_K_M and it is quite impressive.