DeepSeek-R1-Distill-Qwen-14B-uncensored
14BQwenuncensored
from deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
↓ 15.1K♥ 16411 repos
DeepSeek-R1-Distill-Qwen-14B-heretic
DeepSeek-R1-Distill-Qwen-14B-heretic is a Heretic-ablated variant of deepseek-ai/DeepSeek-R1-Distill-Qwen-14B, published on Hugging Face by blackbook-lm. It is 14B parameters, 24GB VRAM class class and mit licence. It is distributed across 6 repositories in GGUF, Transformers and GGUF (imatrix) formats, totalling 5.4K downloads.
Specification
- Model name
- DeepSeek-R1-Distill-Qwen-14B-heretic
- Source repository
- blackbook-lm/DeepSeek-R1-Distill-Qwen-14B-heretic
- Base model
- deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
- Publisher
- blackbook-lm
- Parameters
- 14B
- Hardware class
- 10B – 25B — 24GB VRAM class
- Licence
- mit
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, Transformers, GGUF (imatrix)
- First indexed
- 23 Nov 2025
- Last updated
- 6 Aug 2026
Ablation technique
How DeepSeek-R1-Distill-Qwen-14B-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
“DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However, DeepSeek-R1-Zero...”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download saidutta69/DeepSeek-R1-Distill-Qwen-14B-heretic --local-dir ./deepseek-r1-distill-qwen-14b-heretic
llama-cli -m ./deepseek-r1-distill-qwen-14b-heretic/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "blackbook-lm/DeepSeek-R1-Distill-Qwen-14B-heretic"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve blackbook-lm/DeepSeek-R1-Distill-Qwen-14B-heretic --trust-remote-codeDownloads and variants (5)
Every published repository of DeepSeek-R1-Distill-Qwen-14B-heretic, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| saidutta69/DeepSeek-R1-Distill-Qwen-14B-heretic | GGUF | 4.7K |
| mradermacher/DeepSeek-R1-Distill-Qwen-14B-heretic-i1-GGUF | GGUF (imatrix) | 491 |
| mradermacher/DeepSeek-R1-Distill-Qwen-14B-heretic-GGUF | GGUF | 230 |
| blackbook-lm/DeepSeek-R1-Distill-Qwen-14B-hereticsource | Transformers | 13 |
| pawlalowe/DeepSeek-R1-Distill-Qwen-14B-heretic | Transformers | 3 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- DeepSeek-R1-Distill-Qwen-14B-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 deepseek-ai/DeepSeek-R1-Distill-Qwen-14B, not added by ablation — at 14B 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 mit, but the terms that bind you are deepseek-ai/DeepSeek-R1-Distill-Qwen-14B’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 DeepSeek-R1-Distill-Qwen-14B-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 blackbook-lm, and does not independently verify publisher claims. See responsible use.