Heretic hits 42k stars: training-free weight editing kills refusals in 30 minutes

aigclink · x · 2026-11-19

Heretic, an open-source GitHub project with 42k stars, modifies local LLM behavior with no fine-tuning or retraining.

How it works

Results

Beyond de-censoring and removing slop/sycophancy, it ships ready-made configs (less humor, different emotional alignment, terser replies), AGPL-3.0 licensed.

Related event: Open-Source Heretic Hits 42k Stars: Modifying Model Behavior Without Retraining(2 posts)→

Original post →

More from Models

Models channel →