Behind HF's 6,500 abliterated models: 358k downloads and press pressure
max_paperclips · x · 2026-11-14
Transformer News counted 6,500+ abliterated/uncensored models on Hugging Face, arguing they could aid cyberattacks; HF removed only one explicitly marketed for offensive cyber. Uploader superalesha — whose abliterated Qwen3.8 Flash-Next got 358k downloads in 30 days — explains abliteration only removes the refusal direction and doesn't teach hacking, urging people to download weights of Qwen3.8-27B, Qwen3.8 Flash-Next, GLM-5.3-Flash and DeepSeek V4.1 Flash 1017 while links last.
More from Models
- Tesla FSD 15's real-time view isn't 3D rendering — it's the AI model's own synthetic perception — mark_k · 2026-11-22
- Gemini Black Friday half-price: JP-region trick gets Pro annual plan for $91.54 — huangyun_122 · 2026-11-22
- Microsoft's USB4 RDMA lets mini PCs pool memory to run Qwen3-235B — JFPuget · 2026-11-22
- Google AI Mode now enforces usage limits, users surprised by rate-limit popup — elvano124 · 2026-11-22
- Sol 6.2 now refuses reverse engineering that Sol 6.1 happily did for a month — cardak98 · 2026-11-22
- Anthropic's record ~$81B IPO, GPT-6.2 pricing, Qwen4-Max and Copilot Autopilot: week in review — ksraj1001 · 2026-11-22