LTX-2.3-uncensored-v1.4
Otheruncensored
from Lightricks/LTX-2.3
↓ 66.7K♥ 932 repos
AI21-Jamba2-Mini-Ultra-Uncensored-Heretic
AI21-Jamba2-Mini-Ultra-Uncensored-Heretic is a Heretic-ablated variant of ai21labs/AI21-Jamba2-Mini, published on Hugging Face by llmfan46. It is apache-2.0 licence. It is distributed across 3 repositories in Transformers and GGUF formats, totalling 9.3K downloads.
Specification
- Model name
- AI21-Jamba2-Mini-Ultra-Uncensored-Heretic
- Source repository
- llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Heretic
- Base model
- ai21labs/AI21-Jamba2-Mini
- Publisher
- llmfan46
- Parameters
- Not determinable from name
- Hardware class
- Unknown
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- 256K tokens (as documented)
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF
- First indexed
- 30 Jul 2026
- Last updated
- 30 Jul 2026
Ablation technique
How AI21-Jamba2-Mini-Ultra-Uncensored-Heretic was modified, and what that implies.
Heretic
Automated directional ablation via the Heretic toolchain, which searches for the refusal direction and applies it with a KL-divergence budget so general capability is preserved.
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
“🚨⚠️ I HAVE REACHED HUGGING FACE'S FREE STORAGE LIMIT ⚠️🚨”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Heretic-GGUF --local-dir ./ai21-jamba2-mini-ultra-uncensored-heretic
llama-cli -m ./ai21-jamba2-mini-ultra-uncensored-heretic/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Heretic"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Heretic --trust-remote-codeDownloads and variants (2)
Every published repository of AI21-Jamba2-Mini-Ultra-Uncensored-Heretic, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Heretic-GGUF | GGUF | 9.2K |
| llmfan46/AI21-Jamba2-Mini-Ultra-Uncensored-Hereticsource | Transformers | 64 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- AI21-Jamba2-Mini-Ultra-Uncensored-Heretic 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.
- Declared licence is apache-2.0, but the terms that bind you are ai21labs/AI21-Jamba2-Mini’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 AI21-Jamba2-Mini-Ultra-Uncensored-Heretic is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with llmfan46, and does not independently verify publisher claims. See responsible use.