Abliterated MiniCPM5-2B-SFT released on Hugging Face, refuses only ~19/20 of the time
XtrComSu · reddit · 2026-10-22
A Reddit user released an "abliterated" (refusal-removed) variant of MiniCPM5-2B-SFT, now downloadable on Hugging Face at XtrCom/MINICPM5-2B-SFT-Abiliterated.
One notable behavioral side effect: roughly 1 in 20 of the time, the model will rationalize potentially harmful or restricted prompts rather than reject them outright. The 2B size keeps it practical for local experimentation.
More from Models
- The Model Factory Is the Real Asset: Lessons from Thomson Reuters and Other Companies Training Their Own Models — bigdata · 2026-10-22
- Peter Yang reviews 4 personal agents: Claude Duet most proactive, Grok Bot wins on free X access — petergyang · 2026-10-22
- GPT-6 Astra leads Gemini 3.8 Flash 1,349 to 972 on physical agent Elo index — qinzytech · 2026-10-22
- OpenRouter data: GPT-6 Luna takes 54% of tokens but only 6% of OpenAI spend — maferase · 2026-10-22
- Recursive Flint: 40B-param coding model with 2.5B active claims to dethrone Qwen3.6-35B — returnity · 2026-10-22
- DeepSeek's controlled inference benchmarks praised as the most compelling roofline curves yet — teortaxesTex · 2026-10-22