---
title: 'Interesting Models'
url: 'https://grav.munn.me/ai/interesting-models'
markdown: 'https://grav.munn.me/ai/interesting-models.md'
date: '2025-11-08'
description: "Model Description Rating Qwen3-Next Newest quen3 model which can load ONLY the part you need into each sectoin. None yet Apertus LLM Swiss open sourced model LOTS of languages (written) abliterated qwen Abliteration is the process for removing model safeties. It lets you research 'forbi…"
---

# Interesting Models

| Model | Description | Rating |
|---|---|---|
| [Qwen3-Next](https://x.com/Alibaba_Qwen/status/1966197643904000262) | Newest quen3 model which can load ONLY the part you need into each sectoin. | None yet |
| [Apertus LLM](https://huggingface.co/collections/swiss-ai/apertus-llm-68b699e65415c231ace3b059) | Swiss open sourced model LOTS of languages (written) |  |
| [abliterated qwen](https://huggingface.co/mradermacher/Huihui-Qwen3-30B-A3B-abliterated-Fusion-9010-GGUF) | Abliteration is the process for removing model safeties. It lets you research 'forbidden' topics that the nanny ass llms won't let you look at. | 5 |
| [Lama Uncensored](https://huggingface.co/DavidAU/Llama-3.2-8X3B-MOE-Dark-Champion-Instruct-uncensored-abliterated-18.4B-GGUF) | These ones can be a bit dark, but GREAT for role playing games or fiction. | ? |
| [Granite Guardian](https://huggingface.co/ibm-granite/granite-guardian-3.2-5b-lora-harm-correction) | Can check user prompts for evil | 5 |
|  |  |  |

---

## Navigation

- Parent: [AI](https://grav.munn.me/ai.md)
- Previous: [Progs and Libs](https://grav.munn.me/ai/progs-and-libs.md)