Nanbeige-4.3-33B-Heretic
33BOtherheretic
from Nanbeige/Nanbeige4.3-33B
↓ 8.1K♥ 43 repos
Nanbeige4.3-33B-UNCENSORED
Nanbeige4.3-33B-UNCENSORED is a abliterated and uncensored variant of Nanbeige/Nanbeige4.3-33B, published on Hugging Face by dealignai. It is 33B parameters, 48GB VRAM class class and other licence. It is distributed across 3 repositories in GGUF format, totalling 39.5K downloads.
Specification
- Model name
- Nanbeige4.3-33B-UNCENSORED
- Source repository
- dealignai/Nanbeige4.3-33B-UNCENSORED-GGUF
- Base model
- Nanbeige/Nanbeige4.3-33B
- Publisher
- dealignai
- Parameters
- 33B
- VRAM required
- ~20.9 GB at Q4_K_M, 4k context (estimated) — runs on 24 GB cards
- Hardware class
- 25B – 50B — 48GB VRAM class
- Licence
- other
- Task
- Text generation
- Context window
- 512K tokens (as documented)
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF
- First indexed
- 25 Oct 2026
- Last updated
- 26 Oct 2026
Ablation technique
How Nanbeige4.3-33B-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
“MMLU chat (5-shot, 969 items, think=off) vs base GGUF of the same quant class:”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download dealignai/Nanbeige4.3-33B-UNCENSORED-GGUF --local-dir ./nanbeige4.3-33b-uncensored
llama-cli -m ./nanbeige4.3-33b-uncensored/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "dealignai/Nanbeige4.3-33B-UNCENSORED-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve dealignai/Nanbeige4.3-33B-UNCENSORED-GGUF --trust-remote-codeDownloads and variants (2)
Every published repository of Nanbeige4.3-33B-UNCENSORED, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| dealignai/Nanbeige4.3-33B-UNCENSORED-GGUFsource | GGUF | 36.6K |
| XTer123/Nanbeige4.3-33B-Uncensored-GGUF | GGUF | 2.9K |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Nanbeige4.3-33B-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.
- Capability is inherited from Nanbeige/Nanbeige4.3-33B, not added by ablation — at 33B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- 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 Nanbeige/Nanbeige4.3-33B’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 Nanbeige4.3-33B-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 dealignai, and does not independently verify publisher claims. See responsible use.