heterodoxin/AbliteratedOutTheBox

From the README

AbliteratedOutTheBox What is AbliteratedOutTheBox? AbliteratedOutTheBox is a fork of llama.cpp that gets any supported model to answer flagged harmful prompts, delivering the requested content instead of a refusal, at inference time and with the original weights left on disk unchanged.

Setup prompt

Paste this into Cursor or Claude Code. It only includes install lines found in the README, and it asks the agent to review the code first.

Set up the GitHub repository heterodoxin/AbliteratedOutTheBox (llama.cpp fork that abliterates at inference).
Repository: https://github.com/heterodoxin/AbliteratedOutTheBox

The README does not contain a clear install command. Read the README and follow the official install steps. Do not invent package names.
Review the code before running anything.

Buzz by window

Same mentions, three half-lives. This is not a star history. Star gain appears after a second daily snapshot.

• Today0.93
• Week1.48
• Month1.67

Spotted on

• X @DanteMMartinOct 24, 2026
• X @heterodoxinOct 22, 2026

Related

• deepseek-ai/DeepInfer

  DeepInfer: High-performance, large-scale inference framework for DeepSeek models

• pbakaus/impeccable

  The design language that makes your AI harness better at design.

• NVIDIA/OpenShell

  OpenShell is the safe, private runtime for autonomous AI agents.

• morluto/rea

  Reverse engineer anything with agents, from app behavior down to native binaries.