Heretic Creator Warns: Don't Use 'Heretic' Models as Text Encoders for H3
-p-e-w- · reddit · 2026-08-12
The creator of Heretic issues a PSA advising against using 'heretic' models as text encoders for video generation models like H3. He explains that Heretic uses directional ablation to make LLMs comply, but it doesn't produce more 'raw' representations; it perturbs them, reducing prompt adherence and potentially introducing artifacts. Thus, it won't uncensor video outputs and may degrade quality.
More from Models
- Open Source AI Summer: 13 New Open-Weight Models from DeepSeek, Meta, and More — victormustar · 2026-08-12
- Liquid AI Launches LFM2.5-VL-3B Lightweight Vision Model, Outperforming Larger Rivals — helloiamleonie · 2026-08-12
- LiquidAI Launches LFM2.5-VL-3B for On-Device Multimodal — pmttyji · 2026-08-12
- Feeding Psychedelic Docs to LLMs: Over-Iteration Breeds Eerie Abstractions — ctjlewis · 2026-08-12
- ChatGPT Halts Market Share Decline, Claude Continues Steady Growth — koltregaskes · 2026-08-12
- DeepMind Releases SL2T: Real-Time Sign Language to Text Model — TorturedPoet30 · 2026-08-12