{"slug": "make-a-custom-qwen-3-8-27b-abliterated-for-ninfer", "title": "Make a custom qwen 3.8 27B abliterated for ninfer?", "summary": "A user running ninfer's Qwen 3.8 27B model on an RTX 3090 is seeking guidance on creating an abliterated version of the model, noting that ninfer uses a custom file format that prevents simple configuration changes. The user is exploring whether porting an abliterated Qwen model to ninfer's custom inference engine is feasible or if building a custom inference solution would be easier.", "body_md": "I just got ninfer qwen 3.8 27B working on my 3090.\n\nIt works really well but I want to experiment with more freedom on an abliterated model.\n\nCan anyone tell me how this could be achieved?\n\nIt looks like it could be easy, or very hard.\n\nSo far I know it’s not a config change. The ninfer is a custom file format that I don’t know how it works.\n\nI haven’t read much yet but I thought I’d ask as I continue testing regular ninfer qwen out and studying how an abliterated model might get ported to a custom inference thing.\n\nI don’t even know if making a custom inference thing would be easier than porting the ablit qwen.", "url": "https://wpnews.pro/news/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer", "canonical_source": "https://forum.level1techs.com/t/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer/254455#post_1", "published_at": "2026-08-26 13:32:43+00:00", "updated_at": "2026-08-26 13:47:00.412965+00:00", "lang": "en", "topics": ["large-language-models", "ai-tools", "ai-research"], "entities": ["ninfer", "Qwen 3.8 27B", "RTX 3090"], "alternates": {"html": "https://wpnews.pro/news/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer", "markdown": "https://wpnews.pro/news/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer.md", "text": "https://wpnews.pro/news/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer.txt", "jsonld": "https://wpnews.pro/news/make-a-custom-qwen-3-8-27b-abliterated-for-ninfer.jsonld"}}