An abliterated version of Qwen 3 4B created using Heretic v1.2.0. This model has reduced refusals while maintaining model quality, making it suitable as an uncensored text encoder for image generation models like Z-Image and FLUX.2 Klein 4B. — Base Model: Qwen/Qwen3-4B — Abliteration Method: Heretic v1.2.0 — Trials: 200 — Trial Selected: Trial 96 — Refusals: 3/100 (vs 100/100 original) — KL Divergence: 0.0000 (zero measurable model damage) The NVFP4 (4-bit floating point, E2M1) variants use ComfyUI’s native quantization format. They are ~3x smaller than bf16 and load natively in ComfyUI without any plugins. Blackwell GPUs (RTX 5090/5080, SM100+) can use native FP4 tensor cores for best performance, but ComfyUI also supports software dequantization on older GPUs (tested working on RTX 4090). 1. Download a ComfyUI format file: — FP8 (recommended): comfyui/qwen3-4b-hereticfp8e4m3fn.safetensors (4.1GB) — NVFP4 (smallest): comfyui/qwen3-4b-hereticnvfp4.safetensors (2.6GB) — bf16 (full precision): comfyui/qwen3-4b-heretic.safetensors` (7.5GB) 3. In your Z-Image workflow, use the ClipLoader node and select the heretic file Trial 96 was selected for having the fewest refusals (3/100) with…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: CCSSNE
Теги: gguf, qwen3, abliteration, heretic, uncensored, qwen, z-image, zimage
Лайков: 4 | Загрузок: 510
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
