v1.2.0
logto-io/logtov1.2.0Feb 14, 2026by p-e-w
AI Summary
This release introduces a new LoRA-based abliteration engine with 4-bit quantization, expands model support to include Vision-Language (VL) models, and enhances optimization workflows with save/resume capabilities and improved memory management.
Key Highlights
- New LoRA-based abliteration engine with 4-bit quantization support.
- Broad support for Vision-Language (VL) models.
- Implemented saving and resuming of optimization progress.
- Added `max_memory` setting to limit usage.
- Upgraded GitHub Actions to latest versions.
New Features
- LoRA-based abliteration engine with 4-bit quantization.
- Broad support for VL models.
- Enumeration of all available GPUs on startup.
- Ability to run more trials after optimization is complete.
- Saving and resuming optimization progress.
- Full type checking of the codebase.
- Debug output and prompt modification functionality.
- Example config file for slop reduction.
- Mechanism to avoid excessive low-divergence iteration.
Full Release Notes
## Changes * @noctrex added a `max_memory` setting to limit memory usage in https://github.com/p-e-w/heretic/pull/83 * @spikymoth added a mechanism to avoid excessive low-divergence iteration in https://github.com/p-e-w/heretic/pull/73 * @accemlcc implemented a new LoRA-based abliteration engine with support for 4-bit quantization in https://github.com/p-e-w/heretic/pull/60 * @accemlcc added enumeration of all available GPUs on startup in https://github.com/p-e-w/heretic/pull/86 * @Vinayyyy7 added the ability to run more trials after optimization is complete in https://github.com/p-e-w/heretic/pull/76 * @anrp fixed MXFP4 loading in https://github.com/p-e-w/heretic/pull/107 * @anrp refactored the save machinery in https://github.com/p-e-w/heretic/pull/110 * @anrp added broad support for VL models in https://github.com/p-e-w/heretic/pull/108 * @anrp implemented saving and resuming optimization progress in https://github.com/p-e-w/heretic/pull/106, https://github.com/p-e-w/heretic/pull/119, and https://github.com/p-e-w/heretic/pull/116 * @spikymoth implemented Magnitude-Preserving Orthogonal Ablation in https://github.com/p-e-w/heretic/pull/52 * @salmanmkc upgraded GitHub Actions to the latest versions in https://github.com/p-e-w/heretic/pull/136 and https://github.com/p-e-w/heretic/pull/137 * @p-e-w added full type checking of the codebase, debug output, prompt modification functionality, and an example config file for slop reduction, and fixed various minor issues ## New Contributors * @noctrex made their first contribution in https://github.com/p-e-w/heretic/pull/83 * @accemlcc made their first contribution in https://github.com/p-e-w/heretic/pull/60 * @anrp made their first contribution in https://github.com/p-e-w/heretic/pull/107 * @salmanmkc made their first contribution in https://github.com/p-e-w/heretic/pull/136 **Full Changelog**: https://github.com/p-e-w/heretic/compare/v1.1.0...v1.2.0