{"version":1,"type":"story","url":"https://digestai.news/story/deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4","json":"https://digestai.news/story/deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4.json","markdown":"https://digestai.news/story/deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4.md","slug":"deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4","headline":"DeepSeek releases V4.1 Flash 1017 gguf model with refusals cut to 4%","summary":"DeepSeek has published a new version of its V4.1 Flash LLM, called Flash 1017, in the GGUF format. The uploader applied an \"abliteration\" process that reduces the model’s refusal rate from 97 out of 100 prompts to just 4 out of 100 in chat mode, and from 98 out of 100 to 9 out of 100 in think mode by using a rank‑2 subspace adjustment. \n\nThe GGUF release lets users run the high‑performance model on personal computers or other local environments instead of relying on cloud services, which protects privacy and eliminates the need to send data to external providers. The model and accompanying LoRA adapters are available on HuggingFace, and the contributor emphasizes that the goal is to balance low refusals with strong overall performance rather than forcing refusals to zero.","keyPoints":["Refusal rate drops from 97% to 4% for chat mode after ablation.","Think‑mode refusals fall from 98% to 9% using rank‑2 subspace adjustment.","Model released in GGUF format, runnable locally on PCs, hosted on HuggingFace."],"whyItMatters":"A locally runnable LLM with dramatically lower refusal rates gives developers and small businesses privacy‑preserving AI without cloud costs, expanding practical automation.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["DeepSeek"],"models":["DeepSeek V4.1 Flash 1017"],"people":[]},"firstPublishedAt":"2026-10-18T01:29:00Z","updatedAt":"2026-10-18T01:29:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"note.com","title":"AI unusable due to 'refusals'? [97%→4%] How to run DeepSeek V4.1 locally","url":"https://note.com/tn_world/n/nf1cc357e6134?hl=en","publishedAt":"2026-10-18T01:29:00Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":{"title":"Rapid Rise of Open Weight AI Models","url":"https://digestai.news/thread/hugging-faces-ml-intern-builds-custom-models-from-prompts-in-days","storyCount":4},"cite":{"text":"Digest AI, \"DeepSeek releases V4.1 Flash 1017 gguf model with refusals cut to 4%\", 18 October 2026, https://digestai.news/story/deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4","publisher":"Digest AI","title":"DeepSeek releases V4.1 Flash 1017 gguf model with refusals cut to 4%","datePublished":"2026-10-18T01:29:00Z","url":"https://digestai.news/story/deepseek-releases-v4-1-flash-1017-gguf-model-with-refusals-cut-to-4"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}