mradermacher

/

Trendyol-Cybersecurity-LLM-Qwen3-32B-Q8_0-GGUF-GGUF

like 1

Possible if we have the original model but not useful for Qwen3 based monolithic models. I alliterated XBai-o4 this morning and it is still so censored I haven’t even bothered to upload it. The only way would be to uncensor it using fine-tuning. This is something I could do if the original SafeTensors model would be public. https://huggingface.co/Trendyol/Trendyol-Cybersecurity-LLM-Qwen3-32B-Q8_0-GGUF is a GGUF and not the original model in the SafeTensors format. Quantized GGUFs cannot be finetuned or abliterated.

Thanks for complete explanation.