UNCENSORED AI MODEL NEWS
Find your next model.
Uncensored and abliterated AI model news, cloud GPU costs and self-hosting research. Find a model worth running, then get help deploying it.
Release notes, source links, license details and reported benchmarks. Read the research. Choose your stack. Make it yours.
huihui-ai/Huihui-Qwen3.8-27B-abliterated
Within 48 hours of Qwen3.8-27B's community release, huihui-ai published a 27.8B-parameter abliterated edit that leaves the first 15 layers, MTP and the vision tower untouched, and its GGUF companion landed the same afternoon.
Read the article → Model research ·0bserverx/Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF
Abliteration as a matrix optimization problem, run three times: KL 0.0535 → 0.0085, refusals 3/100 → 0–1/100, 106K downloads in four days, one corrupted quant and a loud community pushback.
Read the article → Model research ·pocharlies/deepseek-v4-flash-0731-uncensored-abliterated-refusal-directions
A 757 KB refusal-directions file turns DeepSeek V4 Flash 0731 abliteration into a runtime λ dial, with measured proof that the baked λ=2.5 checkpoint overshoots and inverts the direction.
Read the article → Model research ·jorkle/Muse-Glimmer-30B-Abliterated
jorkle's KL-measured LoRA-SFT de-refusal of Meta's dense 29.8B agentic model — the Normal twin of a three-repo family with its own GGUF ladder and base-reference quants.
Read the article → Model research ·huihui-ai/Huihui-CyberStrike-OffSec-35B-abliterated
An abliterated fine-tune of the CyberStrike-OffSec-35B offensive-security model: a Qwen3.6-35B-A3B tool-calling base, refusal weights edited out, no post-edit evaluations published.
Read the article → Model research ·huihui-ai/Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated
Qwen architecture, Empero post-training and huihui-ai's final refusal-reduction pass, with a 1M label that deserves a closer look.
Read the article → Model research ·huihui-ai/Huihui-Qwen3.6-35B-A3B-abliterated
The multimodal proof of concept: how huihui-ai applied abliteration to a 36-billion-parameter mixture-of-experts checkpoint, and why uncensored still needs caveats.
Read the article → Model research ·cebeuq/Ornith-1.0-397B-abliterated-W4A16
A shard-by-shard abliteration and W4A16 conversion of a 396.8-billion-parameter sparse model that still produces nearly 196 GiB of weights.
Read the article → Model research ·YuYu1015/YuYu1015-Ornith-1.0-35B-abliterated
DeepReinforce's coding scaffold, YuYu1015's corrected weights and the boundary between upstream and derivative benchmarks.
Read the article →Model research, dated at publication. Model licenses, publisher benchmarks and hosting estimates are specific to each article, not a live availability or price list. Reported zero-refusal results are test-specific, not a universal guarantee.