Lightricks ships a 330MB LTX-2.3 LoRA that removes people from video
Thick_Impression_307 · reddit · 2026-07-22
Lightricks quietly released a Clean Plate IC-LoRA for LTX-2.3-22B that removes people, pedestrians, and vehicles from video without masks or roto.
- The adapter is about 330 MB and is available on Hugging Face and ModelScope.
- It reconstructs the background while preserving camera motion; the tester says it worked on a crowded Madrid street scene.
- Caveats: small sign text can drift, colors may shift, and shots where subjects fill most of the frame do not work well.
- The README recommends describing the empty scene you want, and the post includes a ComfyUI-LTXVideo local workflow tip.
- Realistically, it wants around 24 GB VRAM for HD use.
Related event: LTX-2.3 Clean Plate LoRA Removes People and Cars from Videos Without Masks(7 posts)→
More from coding & agent
- Team-level AI agents: where should shared context and history live? — Al_Grigor · 2026-09-11
- Chaining dependent MCP tool calls: no rollback, duplicate risk — agentrsdg · 2026-09-11
- DeepMind-led paper makes design docs the source of truth, code disposable — SMART regenerates in 1.5-3h for ~$100 — Roger_M_Taylor · 2026-09-11
- Agent-built classifier labels 192k docs for $0.70 vs $13-26 with frontier LLMs — vanstriendaniel · 2026-09-11
- MathModelAgent gains traction: auto-solves math modeling and writes a submission-ready paper — jihe520 · 2026-09-11
- alphaXiv open-sources OpenResearch to run parallel research agents with any model — alphaXiv · 2026-09-11