LTX quietly updates Ingredients IC-LoRA for 2.5: sheet-image-based reference-to-video
kabachuha · reddit · 2026-09-12
Lightricks quietly updated the Ingredients IC-LoRA for LTX-2.5 (22b, now on Hugging Face). The key capability: reference-to-video driven by a single character/scene sheet image, keeping characters and settings consistent across generated video without elaborate prompting. A practical open-source update for consistency-focused video creators.
More from Multimodal
- Midjourney V6.1 style codes: 4 sref IDs with a ready-to-use prompt — michaelrabone · 2026-09-12
- Paper-Tearing Comparison Video Shows a Year of Video-Gen Physics Gains — sabage27 · 2026-09-12
- Removing "AI slop" hallmarks from images with a single prompt to Astra — floguo · 2026-09-12
- Nari Labs open-sources Qwen3-TTS 1.7B: 50ms latency, 10x cheaper than ElevenLabs — alexcovo_eth · 2026-09-12
- Short film 'The Fly' made in 4 hours on one RTX 5090 with ComfyUI, Minimax, Krea 2 and Suno — aurelm · 2026-09-12
- ComfyUI-QwenASR v1.1.0: official Qwen3-ASR models, smart ITN, and long-form forced alignment — Narrow-Particular202 · 2026-09-12