Overview of Uncensored LLM Models
This article presents a selection of uncensored large language models (LLM) hosted on the Hugging Face platform. Given the relative scarcity of recent local model releases, we explore the available options that offer less restrictive behavior.
The listed models include variants of:
- GLM 4.7 Flash
- GPT OSS 20B
- GPT OSS 120B
- Gemma 12B
- Gemma 27B
- Qwen 30B A3B
- Qwen 8B
- Qwen 32B
It is important to note that different models use different methods of removing censorship, which can lead to very different behaviors between them. For those evaluating on-premise deployments, there are trade-offs between control and flexibility.
List of Models and Links
Below is a list of the mentioned models with their respective links to Hugging Face:
- GLM 4.7 Flash:
- GPT OSS 20B:
- https://huggingface.co/DavidAU/OpenAi-GPT-oss-20b-abliterated-uncensored-NEO-Imatrix-gguf
- https://huggingface.co/DavidAU/OpenAi-GPT-oss-20b-HERETIC-uncensored-NEO-Imatrix-gguf
- https://huggingface.co/huihui-ai/Huihui-gpt-oss-20b-BF16-abliterated-v2
- https://huggingface.co/bartowski/p-e-w_gpt-oss-20b-heretic-GGUF
- GPT OSS 120B:
- Gemma 12B:
- Gemma 27B:
- Qwen 30B A3B:
- Qwen 8B:
- Qwen 32B:
💬 Comments (0)
🔒 Log in or register to comment on articles.
No comments yet. Be the first to comment!