// about
About wahloon.net
A focused, independent resource for people running abliterated and uncensored open-weight models locally.
What this site is
wahloon.net tracks abliterated and uncensored open-weight language models. It is built around three things:
- A model catalog. Filterable by family, method, size, VRAM and quantization, with direct links to the original Hugging Face repositories and Ollama tags where they exist.
- Practical guides. Step-by-step setup for Ollama, LM Studio, llama.cpp and sensible quantization choices.
- News and FAQ. Release tracking and clear answers to the questions that come up repeatedly in local-LLM communities.
How the catalog works
Model entries live in a typed data file, data/models.ts. The UI never hardcodes model information; it renders whatever is in that file. News and guides are MDX files in content/news and content/guides. FAQs live in data/faqs.ts.
There is no CMS and no database. That keeps the site fast, cheap to host, and easy to review in a pull request. It also means every published claim has a visible source in the repository.
Editorial approach
Entries link to the original repository. Descriptions are kept short and factual. Download counts, likes and VRAM figures are approximate and change over time; always verify the current model card before downloading.
We do not rank models by “safety” or make claims about legality beyond a neutral, high-level note. Abliteration is a technical method with technical trade-offs. Readers are adults running their own hardware.
Responsibility
Models listed here are open-weight artifacts published by third parties. wahloon.net is not affiliated with the publishers. You are responsible for how you use a model and for complying with its license and applicable law. Generating or distributing illegal content remains illegal regardless of the model.
Machine-readable
Every key page supports Accept: text/markdown. There is also an llms.txt overview, an RSS feed and a sitemap.