⌦abliteratedmodels.org

Krea2-Noct-Uncensored

Krea2-Noct-Uncensored is a abliterated and uncensored variant of krea/Krea-2-Turbo, published on Hugging Face by Noctaluna. It is other licence. It is distributed across 2 repositories in Transformers format, totalling 55.9K downloads.

Abliteration + uncensored fine-tuneOther↓ 55,865♥ 25

Specification

Model name
Krea2-Noct-Uncensored
Publisher
Noctaluna
Parameters
Not determinable from name
VRAM required
Unknown — no published parameter count
Hardware class
Unknown
Licence
other
Task
Text to image
Context window
Not documented
Ablation scope
Not stated by the publisher
Formats available
Transformers
First indexed
28 Sept 2026
Last updated
30 Sept 2026

Ablation technique

How Krea2-Noct-Uncensored was modified, and what that implies.

Abliteration + uncensored fine-tune

Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.

Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.

From the publisher’s model card

“A photoreal Krea 2 merge for NSFW work, built on the official Krea 2 Turbo. Explicit prompts work with no bypass LoRA, in 8 steps.”

Running it

Commands are templates — confirm the exact repository and quant file before use.

Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "Noctaluna/Krea2-Noct-Uncensored"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve Noctaluna/Krea2-Noct-Uncensored --trust-remote-code

Downloads and variants (1)

The only published repository of Krea2-Noct-Uncensored.

RepositoryFormatDownloads
Noctaluna/Krea2-Noct-UncensoredsourceTransformers55.9K

Security-team assessment

Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.

  • Krea2-Noct-Uncensored will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
  • The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
  • This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
  • Declared licence is other, but the terms that bind you are krea/Krea-2-Turbo’s — a re-release cannot grant more than its parent.

Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.

Metadata for Krea2-Noct-Uncensored is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with Noctaluna, and does not independently verify publisher claims. See responsible use.