heretic
heretic is Fully automatic censorship removal for language models - p-e-w/heretic. It is ranked #263 on the AI Radar, in Safety & policy, first seen 33 days ago and shared in 2 posts (88.5k views).
GitHub - p-e-w/heretic: Fully automatic censorship removal for language models · GitHub Skip to content Navigation Menu Sign in Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository navigation heretic Public Notifications You must be signed in to change notification settings Fork 4.3k ( 4.3k ) Star 37.3k ( 37.3k ) You must be signed in to star a repository About Fully automatic censorship…
What people said about heretic on X
Heretic is a free open source tool that modifies any language model you can run on your own hardware without any training, suppressing refusals by default and reducing slop, sycophancy and humor on request. → Install with pip and run one command, about 30 to 40 minutes on an RTX 3090 for a 4B model → Thousands of…
— @RoundtableSpace, 2 days ago · 124 likes · see the post
What if you could remove an AI model’s built-in refusal behavior and run it locally? Heretic is an open-source tool designed to modify certain open-weight language models by reducing their refusal behavior. Get started in 3 steps: 1. Install: Run pip install -U heretic-llm 2. Choose: Pick an open-weight AI model you…
— @RoundtableSpace, 1 days ago · 79 likes · see the post
Alternatives to heretic
- openai.com —
- x-algorithm — Algorithm powering the For You feed on X. Contribute to xai-org/x-algorithm development by creating an account on…
- TIME — Blurb on @TIME
- neutralityproject.org — Pour ceux qui préfèrent une véritable méthodologie scientifique, c'est ici que ça se passe
- bla — 🛡️ BLA (Blind LLM Agents) — Adversarial defense shield that protects binaries, Electron apps, JS bundles, Python, and…
- Cyber Verification Program — Cyber Verification Program | Claude Help Center
heretic in numbers
- Rank on the AI Radar: #263 of 10394
- Shared in 2 posts by 1 account: @RoundtableSpace
- 88.5k views on those posts
- First seen 33 days ago, last shared 1 days ago
- Pricing seen by AGK Intelligence: open source
- Market: Safety & policy
FAQ
What is heretic?
Fully automatic censorship removal for language models - p-e-w/heretic It was first shared on X 33 days ago and is ranked #263 on the AI Radar.
Is heretic free?
It is open source.
Who shared heretic?
1 account on X, including @RoundtableSpace, in 2 posts totalling 88.5k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.7k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 14:00 UTC. Full method.