Run an uncensored AI agent locally

To run an uncensored AI agent locally, serve an abliterated open-weight model behind an OpenAI-compatible server on your machine and point the framework at it. Qwen 3.8 27B Abliterated fits one 24 GB GPU or a 32 GB Mac at 4-bit.

Updated November 18, 2026By unrestrictedaiagent

Models

ModelParametersSize at 4-bitHardwareLicense
Qwen 3.8 27B Abliterated27.8Babout 17 GBone 24 GB GPU or a 32 GB MacApache 2.0
Qwen 3.6 35B-A3B Abliterated36B total, 3B activeabout 22 GBone 24 GB GPUApache 2.0
Gemma 4 31B Heretic31.3Babout 19 GBone 24 GB GPUApache 2.0
Llama 3.1 8B Abliterated8Babout 5 GBan 8 GB GPU or a laptopLlama 3.1 Community License

Sizes at 4-bit are estimates. Models read on Hugging Face November 18, 2026.

Framework settings

FrameworkLocal settingSource
browser-useChatOpenAI(model, base_url="http://localhost:11434/v1", api_key="ollama")README
Hermes Agentbase_url: http://localhost:11434/v1Providers
GooseBuilt-in Ollama and LM Studio providersProviders
Letta/connect, then Ollama or LM StudioModels
smolagentsOpenAIModel(api_base="http://localhost:11434/v1")Models

Local or hosted

LocalHosted
Parameters8B to 36B320B to 2,780B
ContextLimited by your memory1,048,576 tokens
HardwareA local GPU or MacNone
CostHardware and powerFrom $0.24 per 1M input tokens

Hosted figures are for the models on this site.