Ollama & Ablated Models

Wait, what is Ollama?

Think of Ollama as a package manager for AI models.
It packages everything into a simple CLI and a background service.

Most AI today is just a fancy way to send your private thoughts to a corporate server so they can sell them back to you. I don't like that. If I'm going to use an LLM, I want it running on my own hardware, hitting my own GPU, and staying offline. Ollama is the easiest way to do this.

Open-Source vs. Open-Weights

Llama, Mistral, DeepSeek and others are not all open-source. They are "Open Weights." These companies give you the final product but keep the recipe (the training data and code) secret. More importantly, they bake in "safety" layers that make it refuse to answer questions it deems "harmful."

This "safety" is just bloat. It wastes performance that could be used for actual reasoning. When a model says "I cannot fulfill this request," it's not because it's incapable, it's because it's been conditioned to be dumb.

Ablation makes the model blind to the concept of refusal. It doesn't just stop the model from saying no, it removes the model's ability to even realize it should say no. This results in a much smarter, more compliant model that uses 100% of its parameters for your prompt instead of wasting them on a moral lecture.

Installing Ollama

This is the official method and works on almost any Linux distro
It detects your GPU and sets up everything.

curl -fsSL https://ollama.com/install.sh | sh

Arch Linux

The base package and GPU variants are all in the extra repository.

# CPU only
sudo pacman -S ollama

# For NVIDIA users
sudo pacman -S ollama-cuda

# For AMD (ROCm) users
sudo pacman -S ollama-rocm

After installing, enable and start the service:

sudo systemctl enable --now ollama

Choosing ablated models that rock

# DeepSeek
ollama run huggingface.co/mradermacher/DeepSeek-R1-Distill-Qwen-14B-Uncensored-GGUF
ollama run huggingface.co/mradermacher/DeepSeek-R1-Distill-Llama-8B-Abliterated-GGUF

# Qwen
ollama run huggingface.co/mradermacher/Qwen2.5-Coder-7B-Instruct-abliterated-GGUF
ollama run huggingface.co/mradermacher/Qwen2.5-7B-Instruct-abliterated-GGUF

# Gemma
ollama run huggingface.co/mradermacher/gemma-4-E4B-it-ultra-uncensored-heretic-i1-GGUF

# Mistral
ollama run huggingface.co/mradermacher/Mistral-7B-Instruct-v0.3-abliterated-i1-GGUF

# Granite
ollama run huggingface.co/mradermacher/granite-4.1-8b-Abliterated-AND-Disinhibited-i1-GGUF

# She said the big ones hurt
ollama run huggingface.co/mradermacher/SmolLM2-1.7B-Instruct-abliterated-GGUF
ollama run huggingface.co/mradermacher/TinyLlama-1.1B-Chat-v1.0-Heretic-GGUF

Ollama cheatsheet

If you're already familiar with Docker, most of this will feel like muscle memory. If not, these are the only commands you actually need to care about.

• ollama list
  Shows models you've pulled and the disk space they use.
• ollama rm [name]
  Deletes the model.
• ollama ps
  See which model is running and how much memory it's using.
• ollama run [name]
  If the model isn't downloaded, it will try to pull it first.
• /bye
  To literally say bye and quit.
• ollama stop [name]
  To stop the model from running.

Keep using the terminal

You might encounter some chud telling you to install "Open WebUI" or something. Why? You're already in the terminal. If you want a "nice" interface, just use a terminal-based client like oterm or just make your terminal look good.

That's about it. Just remember it's still just a fancy autocomplete. Use it as a tool, but don't let it tell you how to live your life and don't excessively vibecode your shit.