29 skills found
PurpleDoubleD / Locally UncensoredPlug-and-play local AI studio: uncensored chat, image & video generation, coding agent. Runs abliterated LLMs + ComfyUI 100% offline. One installer, no Docker, no cloud.
Orion-zhen / AbliterationMake abliterated models with transformers, easy and fast
pmarreck / Gemma4 HereticalAbliterated Gemma 4 31B for Ollama and MLX — one-command setup with correct chat template fix
AEON-7 / Comfyui Aeon SparkBleeding-edge ComfyUI for NVIDIA DGX Spark (GB10/Blackwell/sm_121a). CUDA 13 + SageAttention v3 (sm_121a) + NVFP4 + 14 custom-node packs + Flux 2 Dev / LTX 2.3 22B / ACE-Step v1.5 XL Turbo pre-bundled with abliterated text-encoder paths.
AEON-7 / Ornith 1.0 35B AEON Ultimate UncensoredUncensored/abliterated Ornith-1.0-35B (AEON Ultimate): 0% refusal, 0 coding-capability loss. BF16 + FP8 for vLLM.
drowzeys / DeepSeek V4 Flash DSpark Abliterated Uncensored 1M 57toksDeepSeek-V4-Flash-DSpark abliterated (uncensored) · ~100% refusal bypass · C1 ~57 tok/s · 1M ctx · 2× DGX Spark · HF weights
MiaAI-Lab / DeepSeek V4 Flash DSpark Abliterated UncensoredDeepSeek V4 Flash DSpark Abliterated (Uncensored) — 2× DGX Spark serving
hec-ovi / Llama Vulkan StrixDocker Compose for llama.cpp GGUF servers on AMD Strix Halo: Qwen, Gemma, and Laguna packages (abliterated and quantized), stock Vulkan plus ROCmFP4/MTP and ROCmFPX, parallel slots, with prefill/decode and quality metrics measured on this rig.
199-biotechnologies / Gemma 4 AbliteratedGemma 4 31B Abliterated — quality-preserving guardrail removal for Google's most capable open model. Apache 2.0. Runs on Apple Silicon via MLX.
kabachuha / Abliterate.cppAbliterate large quantized models at home using llama.cpp!
lna-lab / Gemma4 12b Vllm Sm120Reproducible recipe: serve abliterated Gemma-4-12B (gemma4_unified) at 50-118 tok/s on no-NVLink Blackwell (SM120) via vLLM nightly + ModelOpt FP8/NVFP4 + MTP spec-decode.
DevCop95 / CYHBeriteratusAn uncensored local LLM interface built for security engineers. It enables private, filter-free interaction with abliterated models for vulnerability analysis and advanced scripting without cloud-based restrictions.
drowzeys / Keys GLM5.2 Quantrio INT4 INT8 Mixed Abliterated C1 30toks 4x DGX SparksGLM-5.2 Quantrio INT4/INT8 Mixed Abliterated — SPEED=1 C1≈30 tok/s @ 128k on 4x DGX Spark. Step-by-step recipe, image bake, results.
tonyd2wild / DeepSeek V4 Flash DSpark Abliterated Uncensored 2x DGX SparkKeys/drowzeys' abliterated (uncensored) DeepSeek-V4-Flash-DSpark served on 2x DGX Spark TP=2. Fork of the DS4 DSpark recipe with the model swapped in. Full model credit: Keys/drowzeys.
drowzeys / Keys Latest GLM 5.2 Quantrio INT4 INT8 Mixed Abliterated DFlash 4x DGX Sparkskeys-latest: GLM-5.2 QuantTrio INT4/INT8 ablit + DFlash · 4x DGX Spark · k=12 · 128k · seqs=4 · KV 10GiB
Jumborombo / NSrenFWA Python GUI app that uses your locally-hosted **Qwen2.5-VL-7B-NSFW-Caption-V3-abliterated** model via Ollama to analyze adult videos and rename them with explicit, vulgar, content-aware descriptions.
AEON-7 / Supergemma4 26b Abliterated Multimodal Nvfp4NVFP4 AWQ Full quantization of SuperGemma4-26B-Abliterated-Multimodal for Blackwell GPUs — pre-built vLLM container + patches included
AEON-7 / Nemotron 3 Nano Omni AEON Ultimate UncensoredNemotron-3-Nano-Omni AEON Ultimate Uncensored: 12-D abliterated multimodal reasoning model (BF16 + NVFP4) for DGX Spark / Blackwell. Source-built vLLM v0.20.0 image + 4 patches + bench + DGX Spark deployment guide.
PRITHIVSAKTHIUR / Auto Abliterationmodify a language model's behavior by abliterating its weights.
tonyd2wild / GLM 5.2 Abliterated NVFP4 316K 4x DGX SparkAbliterated GLM-5.2 with a true 4-bit NVFP4 KV cache on 4x DGX Spark: 316K context, 317,279-token pool (+58.6% vs fp8), 41.4 tok/s peak — with honest content-dependent decode measurements.
