LTX-2.5 Layout to Render turns a 3D blockout into a finished shot
Lightricks released an open IC-LoRA for its LTX-2.5 22B video model that turns a grey clay viewport or blocky playblast from Blender or Unreal into a finished shot. The layout sets the camera, motion and object placement, and an art-directed first frame sets the look.
What it does
You give it two things: a layout video (a clay render or simple shapes carrying the camera move) and a first frame you've art-directed from that layout's first frame. It returns a video that moves like the layout and looks like the first frame, with materials, lighting and palette taken from the image.
It's a rank-128 IC-LoRA on LTX-2.5-22B, used with a dedicated two-stage distilled ComfyUI workflow or Lightricks' Python pipeline.
Results and limits
The model card and docs give setup and recommended settings but no measured results or benchmarks. Both the layout video and the first frame are required; the docs say the graph connects them with no bypass.
Worth knowing
Recommended settings are 1920×1088, 24 fps, a frame count of 8k+1 and LoRA strength 1.0, using the LTX-2.5 distilled stack with the 2× latent upscaler. The Hugging Face repo is gated and uses the LTX-2.x Community License. The repo was created on September 22 and last updated on October 1, 2026.
Try it yourself
Export a grey playblast of [your Blender shot], paint or generate an art-directed version of its first frame, then in the LTX-2.5 Layout to Render workflow load the clay video plus that frame and prompt: "A finished cinematic render with the materials, lighting, and palette of the reference image."