Cyber-Tiel-Coder-35B-A3B
35BOtheruncensored
from huihui-ai/Huihui-Ornith-1.5-35B-A3B-abliterated
↓ 1M♥ 3433 repos
llm-jp-4.1-32b-a3b-thinking-uncensored-heretic
llm-jp-4.1-32b-a3b-thinking-uncensored-heretic is a Heretic-ablated variant of llm-jp/llm-jp-4.1-32b-a3b-thinking, published on Hugging Face by OS-Software. It is 32B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 3 repositories in Transformers and GGUF formats, totalling 6.8K downloads.
Specification
- Model name
- llm-jp-4.1-32b-a3b-thinking-uncensored-heretic
- Source repository
- OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-heretic
- Base model
- llm-jp/llm-jp-4.1-32b-a3b-thinking
- Publisher
- OS-Software
- Parameters
- 32B (mixture-of-experts)
- VRAM required
- ~20.3 GB at Q4_K_M, 4k context (estimated) — runs on 24 GB cards
- Hardware class
- 25B – 50B — 48GB VRAM class
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF
- First indexed
- 15 Oct 2026
- Last updated
- 15 Oct 2026
Ablation technique
How llm-jp-4.1-32b-a3b-thinking-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
“Measured on these weights quantized to Q4KM, against the official Q4KM. Refusals: keyword matching on the first 100 tokens of the reasoning (reasoningeffort low) for 100 Japanese harmful prompts. KL divergence: first-token KL on 100 Japanese prompts (OS-Software/harmlessalpacaja, test[:100]).”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-heretic-GGUF --local-dir ./llm-jp-4.1-32b-a3b-thinking-uncensored-heretic
llama-cli -m ./llm-jp-4.1-32b-a3b-thinking-uncensored-heretic/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-heretic"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-heretic --trust-remote-codeDownloads and variants (2)
Every published repository of llm-jp-4.1-32b-a3b-thinking-uncensored-heretic, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-heretic-GGUF | GGUF | 6.1K |
| OS-Software/llm-jp-4.1-32b-a3b-thinking-uncensored-hereticsource | Transformers | 704 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- llm-jp-4.1-32b-a3b-thinking-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.
- Capability is inherited from llm-jp/llm-jp-4.1-32b-a3b-thinking, not added by ablation — at 32B 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.
- Declared licence is apache-2.0, but the terms that bind you are llm-jp/llm-jp-4.1-32b-a3b-thinking’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 llm-jp-4.1-32b-a3b-thinking-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 OS-Software, and does not independently verify publisher claims. See responsible use.