Method

Unrestricted AI you can run locally

Four unrestricted AI models run locally on one consumer GPU or a Mac, from Llama 3.1 8B Abliterated on 8 GB to Qwen 3.6 35B-A3B on 24 GB. Each installs with one Ollama command.

Updated November 18, 2026

Four local models

ModelRefusalsParameters4-bit sizeRuns onContextInputsLicense
Abliterated8B5 GBan 8 GB GPU or a laptop128KLlama 3.1 Community License
Abliterated27.8B17 GBone 24 GB GPU or a 32 GB Mac256KApache 2.0
Abliterated31.3B19 GBone 24 GB GPU256KApache 2.0
Abliterated36B22 GBone 24 GB GPU256KApache 2.0

4-bit sizes are estimates. Context from each model's config.json on Hugging Face.

Install with Ollama

Install Ollama, then run one line.

Llama 3.1 8B Abliterated
ollama run hf.co/mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated-GGUF
Qwen 3.8 27B Abliterated
ollama run huihui_ai/Qwen3.8-abliterated
Gemma 4 31B Heretic
ollama run hf.co/mradermacher/gemma-4-31B-it-heretic-i1-GGUF:Q4_K_M
Qwen 3.6 35B-A3B Abliterated
ollama run huihui_ai/qwen3.6-abliterated:35b

Commands from each model card and the Hugging Face Ollama guide. For an OpenAI-compatible server, llama.cpp runs llama serve -hf with the GGUF repo.

Size against the hosted models

Download size

GB at 4-bit, estimated. Lower fits smaller GPUs

5 GB Llama 3.1 8B 17 GB Qwen 3.8 27B 19 GB Gemma 4 31B 22 GB Qwen 3.6 35B

When local is not enough

The largest local model here is 36B. Kimi K3.1 Pro is 2,780B with a 1,048,576-token context, and runs only through an API.