Uncensored LLMs
227 uncensored language models, most downloaded first: the model each is based on, how its limits were removed, and where to get it.
AIModelFind lists these models from public sources; it does not host or run them. Check each model's license and the law where you use it.
| Model | Based on | Method | Download as | Downloads / mo | Hosted API |
|---|---|---|---|---|---|
 Huihui-Qwen3.6-35B-A3B-abliterated by huihui-ai | Qwen3.6 35B-A3B | Abliterated | FP8 · safetensors | 72K | Weights only |
| DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking | Abliterated | GGUF | 70K | Weights only | |
l Qwen3.5-9B-abliterated by lukey03 | Qwen3.5 9B | Abliterated | GGUF | 70K | Weights only |
Huihui-CyberStrike-OffSec-35B-abliterated by huihui-ai | oyildirim/CyberStrike-OffSec-35B | Abliterated | GGUF | 66K | Weights only |
R Dolphin3-Cyber-8B by RavichandranJ | — | Fine-tuned | GGUF | 61K | Weights only |
Huihui-Qwen3.6-27B-abliterated-MTP by huihui-ai | Qwen3.6 27B | Abliterated | GGUF | 61K | Weights only |
c Qwen3-VL-32B-Instruct-heretic-v2 by coder3101 | Qwen3-VL-32B-Instruct | Abliterated | safetensors | 59K | Weights only |
| Qwen3.8 27B | Fine-tuned | GGUF | 59K | Weights only | |
H Qwen3.6-27B-Uncensored-HauhauCS-Aggressive by HauhauCS | Qwen3.6 27B | Fine-tuned | safetensors | 57K | Weights only |
Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated by huihui-ai | empero-ai/Qwythos-9B-Claude-Mythos-5-1M | Abliterated | GGUF | 56K | Weights only |
| Mistral Small 3 | Fine-tuned | GGUF · safetensors | 54K | Weights only | |
f | Llama 3 70B Instruct | Abliterated | GGUF | 53K | Weights only |
Llama-3.3-70B-Instruct-abliterated by huihui-ai | Llama-3.3-70B-Instruct | Abliterated | GGUF | 51K | Weights only |
 Qwen3.8-Flash-Next-UNCENSORED by dealignai | Qwen3.8 Flash Next | Fine-tuned | GGUF | 50K | Weights only |
H Ternary-Bonsai-2-27B-Abliterated by Hikari07jp | prism-ml/Ternary-Bonsai-2-27B-gguf | Abliterated | GGUF | 48K | Weights only |
Qwen2.5-Coder-14B-Instruct-abliterated by huihui-ai | huihui-ai/Qwen2.5-Coder-14B-Instruct-abliterated | Abliterated | GGUF | 48K | Weights only |
Ornith-1.5-35B-A3B-UNCENSORED by dealignai | ornith-ai/Ornith-1.5-35B-A3B | Fine-tuned | GGUF | 46K | Weights only |
B Ternary-Bonsai-2-27B-Abliterated-PTQ1_0 by BoldingBuilds | prism-ml/Ternary-Bonsai-2-27B-gguf | Abliterated | GGUF | 46K | Weights only |
w Qwen3.8-Flash-Next-Abliterated by windowsxp811203 | Qwen3.8 Flash Next | Abliterated | GGUF | 44K | Weights only |
B Qwen3.8-27B-ABLITERATED by Blackfrost-AI | Qwen3.8 27B | Abliterated | GGUF | 44K | Weights only |
| Gemma 4 12B Instruct | Abliterated | GGUF | 43K | Weights only | |
l Qwen3.6-35B-A3B-uncensored-heretic by llmfan46 | Qwen3.6 35B-A3B | Abliterated | GGUF · GPTQ | 43K | Weights only |
| Qwen3.8 27B | Abliterated | GGUF | 43K | Weights only | |
Qwen2.5-Coder-7B-Instruct-abliterated by huihui-ai | Qwen2.5-Coder-7B | Abliterated | GGUF | 43K | Weights only |
c Llama-3.2-3B-Instruct-uncensored by chuanli11 | Llama 3.2 3B Instruct | Fine-tuned | GGUF | 42K | Weights only |
Huihui-GLM-5.3-Flash-abliterated by huihui-ai | GLM-5.3-Flash | Abliterated | GGUF | 42K | Weights only |
p | Qwen3-VL-32B-Instruct | Abliterated | GGUF | 41K | Weights only |
B Ternary-Bonsai-2-27B-Abliterated-v2-PQ2_0-MTP by BoldingBuilds | prism-ml/Ternary-Bonsai-2-27B-gguf | Abliterated | GGUF | 40K | Weights only |
Huihui-Qwen3-30B-A3B-Instruct-2507-abliterated by huihui-ai | Qwen3 30B-A3B Instruct 2507 | Abliterated | — | 40K | Weights only |
z Qwen2.5-72B-Instruct-abliterated by zetasepic | Qwen2.5-72B-Instruct | Abliterated | GGUF | 40K | Weights only |
0 Qwen3.8-Flash-Next-Uncensored-GSQ-RCO-12-GB-GPU by 0xSojalSec | Qwen3.8 Flash Next | Fine-tuned | safetensors | 40K | Weights only |
| ornith-ai/Ornith-1.5-35B-A3B | Abliterated | GGUF | 39K | Weights only | |
Huihui-Qwen3-VL-4B-Instruct-abliterated by huihui-ai | huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated | Abliterated | GGUF | 38K | Weights only |
| DavidAU/Llama-3.2-8X3B-MOE-Dark-Champion-Instruct-uncensored-abliterated-18.4B | Abliterated | GGUF | 37K | Weights only | |
a DarkIdol-Llama-3.1-8B-Instruct-1.2-Uncensored by aifeifei798 | Llama 3.1 8B Instruct | Fine-tuned | GGUF | 37K | Weights only |
H Qwen3.6-27B-Uncensored-HauhauCS-Balanced by HauhauCS | Qwen3.6 27B | Fine-tuned | safetensors | 37K | Weights only |
Huihui-gpt-oss-20b-BF16-abliterated by huihui-ai | GPT OSS 20B | Abliterated | GGUF | 36K | Weights only |
Huihui-gemma-4-26B-A4B-it-abliterated by huihui-ai | Gemma 4 26B A4B IT | Abliterated | GGUF | 36K | Weights only |
Huihui-Agents-A1-abliterated by huihui-ai | InternScience/Agents-A1 | Abliterated | GGUF | 36K | Weights only |
O Ternary-Bonsai-2-27B-Uncensored-Heretic-v2 by OS-Software | prism-ml/Ternary-Bonsai-2-27B-gguf | Abliterated | GGUF | 35K | Weights only |
| DeepSeek V4 Flash 0731 | Abliterated | safetensors | 35K | Weights only | |
Huihui-DeepSeek-V4-Flash-Vision-Exp-abliterated by huihui-ai | DeepSeek V4 Flash Vision Exp | Abliterated | GGUF | 35K | Weights only |
a Qwen3.8-27B-RANA-abliterated by abliterant | Qwen3.8 27B | Abliterated | GGUF | 35K | Weights only |
Qwen3-14B-abliterated by huihui-ai | Qwen3 14B | Abliterated | GGUF | 34K | Weights only |
b Qwen2.5-14B_Uncensored_Instruct by bartowski | Qwen2p5 14B | Fine-tuned | GGUF | 34K | Weights only |
N Qwen3.8-9B-heretic-uncensored by Noobito45 | empero-ai/Qwen3.8-9B-Distill | Abliterated | GGUF | 34K | Weights only |
d Dongfangshuo-Qwen3.8-27B-NSFW by dongfangshuo | Qwen3.8 27B | Fine-tuned | GGUF · safetensors | 34K | Weights only |
e | OS-Software/Ternary-Bonsai-2-27B-Uncensored-Heretic-GGUF | Abliterated | safetensors | 33K | Weights only |
G GLM-5.3-Flash-Uncensored-RCO-GSQ by GCSA-AiLab | GLM-5.3-Flash | Fine-tuned | GGUF | 33K | Weights only |
 gpt-oss-20b-Derestricted by ArliAI | GPT OSS 20B | Fine-tuned | GGUF | 33K | Weights only |
Frequently asked questions
A model with the refusals and content filters of the original taken out: by removing the refusal direction from its weights (abliteration), by fine-tuning on unfiltered data, or by merging models. What it knows stays the same; what it agrees to answer changes.
No. AIModelFind collects public information only: who released each model, what it is based on, where its weights are and which providers host it. The downloads and APIs belong to their authors and providers.
Most come as GGUF or MLX files, which run locally in tools such as llama.cpp, Ollama or LM Studio. Pick a quantization that fits your memory; the larger the file, the closer it stays to the full model.
Can't find what you're looking for? Contact us