Vectors / Methods

V6: Fine-tune imitation

A vector that approximates the output shift of an existing fine-tune or LoRA, so the effect can be applied without loading its weights.

Original source: No canonical paper; method described by the authors in their READMEs

Repos

1

Vectors

30

All repos using V6

andyoneal/ecce-vectors

V6 Gemma · other · latest commit 2026-10-05

Source cited: ecce-vectors fitting code (MLX) and GUIDE.md included in the repo; llama.cpp GGUF control vectors

30 vectors 30 sets · 36.4 MB
862 repo downloads

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog