IC-LoRA Adapters | LTX Documentation

This page documents the IC-LoRAs trained by LTX. IC-LoRAs use in-context conditioning to perform specialized tasks on video, from structural control and motion guidance to VFX, video restoration, and creative transforms. Each adapter is trained for a specific task.

For an introduction to how IC-LoRAs work, see the IC-LoRA Guide.

Control

IC-LoRAs that guide generation using conditioning signals from a reference — structural signals like depth maps, edge contours, pose skeletons, or motion trajectories, as well as reference sheets that maintain character and object consistency.

Union Control

A single IC-LoRA that handles multiple control types (depth, canny, and pose) in one checkpoint.

Model: LTX-2.3-22b-IC-LoRA-Union-Control

Compatibility: LTX-2.3

Use cases:

Apply depth, canny, and pose control from a single model without swapping IC-LoRA checkpoints between runs
Simplify ComfyUI workflows by replacing three separate IC-LoRA models with one unified control adapter
Reduce VRAM footprint when working with multiple control types by keeping a single IC-LoRA in memory

Workflow: ComfyUI Union Control workflow

Motion Control

Guides object motion using sparse spline-based trajectories rendered as trails of circles. Supports single or multiple simultaneous motion paths for directing object movement within a scene.

Model: LTX-2.3-22b-IC-LoRA-Motion-Track-Control

Compatibility: LTX-2.3

Use cases:

Direct object paths in product showcases, character scenes, or nature shots by drawing spline trajectories
Create intentional, controlled object motion that is more precise than description via text
Guide multiple objects simultaneously along independent paths within a single generation

Key nodes: This IC-LoRA uses two dedicated ComfyUI nodes for defining motion paths:

LTX Draw Tracks — Visual canvas for drawing spline-based motion trajectories
LTX Sparse Track Editor — Fine-tune keypoint positions and timing, converts sparse keypoints into interpolated per-frame coordinate paths

See Sparse Track Conditioning in the IC-LoRA Guide for best practices on preparing motion paths.

Workflow: ComfyUI Motion Control workflow

Ingredients

Conditions video generation on a reference sheet: a single composite image showing characters, props, and locations laid out on a black background. Generated videos keep those elements visually consistent. You provide the reference sheet as a static video and a two-part prompt (Reference sheet: <panels> / Generated video: <action>); the model renders a clip whose characters, props, and setting match the sheet.

Model: LTX-2.3-22b-IC-LoRA-Ingredients

Compatibility: LTX-2.3

Use cases:

Maintain character consistency (face, costume, body) across multiple generated clips using a shared reference sheet
Lock props and location details to a predefined visual inventory for scene-to-scene continuity
Generate action sequences where recurring characters stay on-model without per-frame manual control

Workflow: IC-LoRA Video Input

VFX

IC-LoRAs that add visual effects to existing footage, re-rendering a reference clip with new elements while preserving subject identity, framing, and camera movement.

Water Simulation

Adds water to a reference clip (rivers, surf, rain, waterfalls, floods, splashes, spray, and wet-surface reflections) while keeping the subject’s identity, clothing, pose, camera framing, and background geometry identical to the reference. You provide a “dry” clip and a prompt describing the water you want; the model re-renders the same shot with water added in natural motion.

Model: LTX-2.3-22b-IC-LoRA-Water-Simulation

Compatibility: LTX-2.3

Use cases:

Add realistic water interaction to live-action footage without camera tracking, 3D modeling, or fluid simulation
Generate floods, rain, surf, or splashing water that interacts naturally with moving subjects
Create wet-surface effects (reflections, specularities, darkened materials) that match the existing lighting

Workflow: IC-LoRA Video Input

Utility

IC-LoRAs that restore, repair, or enhance existing video: removing artifacts, filling in missing regions, or improving quality while maintaining the subject or composition.

Colorization

Restores natural color to grayscale, monochrome, or desaturated video while preserving subject identity, framing, and geometry. Only the color information changes.

Model: LTX-2.3-22b-IC-LoRA-Colorization

Compatibility: LTX-2.3

Use cases:

Colorize black-and-white or monochrome archival footage
Restore color to heavily desaturated video
Add plausible natural color to grayscale renders or previsualization footage

Workflow: IC-LoRA Video Input

Decompression

Removes compression artifacts from low-bitrate video (macroblocking, chroma bleed, ringing, and banding) while preserving subject identity, framing, and geometry.

Model: LTX-2.3-22b-IC-LoRA-Decompression

Compatibility: LTX-2.3

Use cases:

Clean up heavily compressed footage for reuse in higher-quality productions
Remove visible macroblocking and ringing from low-bitrate source material
Restore sharp detail and clean edges to video degraded by aggressive encoding

Workflow: IC-LoRA Video Input

Deblurring

Restores sharpness to out-of-focus or defocused video while preserving subject identity, framing, and scene geometry. Only focus and sharpness change.

NOTE: Targets spatial defocus blur specifically; not designed for motion blur, noise, or compression artifacts.

Model: LTX-2.3-22b-IC-LoRA-Deblur

Compatibility: LTX-2.3

Use cases:

Recover sharpness from genuinely out-of-focus footage across people, wildlife, nature, cities, food, and night scenes
Restore crisp detail and clean edges to softly defocused video without altering composition
Sharpen footage shot with incorrect focus without reshooting

Workflow: IC-LoRA Video Input

Inpainting / Outpainting

Extends the canvas of existing video (outpainting) or fills masked regions within video frames (inpainting), generating new content that blends seamlessly with the surrounding scene. Uses mask-aware preprocessing and Laplacian pyramid blending at both generation stages for clean integration.

Model: LTX-2.3-22b-IC-LoRA-In-Outpainting

Compatibility: LTX-2.3

Use cases:

Extend video beyond its original frame boundaries to create wider or taller compositions
Fill masked regions of video with generated content that matches the surrounding scene
Change aspect ratios of existing footage by generating the missing areas

For setup, workflow details, and node documentation, see the In-Outpainting guide.

Creative

IC-LoRAs that apply creative transforms to video, such as changing lighting conditions, modifying appearance, or applying stylistic changes while preserving the underlying scene structure.

Day to Night

Re-renders a daytime video as the same shot at night while preserving composition, framing, camera movement, and subject motion frame-for-frame. The prompt steers the style of night (moonlight, color temperature, brightness) while the reference dictates structure and movement.

Model: LTX-2.3-22b-IC-LoRA-Day-To-Night

Compatibility: LTX-2.3

Use cases:

Convert outdoor daytime footage to photorealistic nighttime without reshooting
Relight street scenes, landscapes, or exterior shots for night sequences
Experiment with different night styles (warm tungsten, cool moonlight, urban neon) from the same daytime reference

Workflow: IC-LoRA Video Input

Instant Shave

Removes beards, mustaches, and stubble from video while preserving the subject’s identity, expression, motion, lighting, and the surrounding scene. You provide a clip of a bearded subject and the model re-renders it clean-shaven.

Model: LTX-2.3-22b-IC-LoRA-Instant-Shave

Compatibility: LTX-2.3

Use cases:

Remove beards, mustaches, or stubble from portrait and full-body video clips
Create clean-shaven versions of footage without reshooting
Preview different looks on the same subject by toggling facial hair on and off

Workflow: IC-LoRA Video Input

Cross-Eyed

Transforms close-up portrait videos from straight eyes to convergent strabismus (crossed eyes), while preserving facial expression, head movement, lighting, and camera framing.

Model: LTX-2.3-22b-IC-LoRA-Cross-Eyed

Compatibility: LTX-2.3

Use cases:

Apply a crossed-eyes effect to portrait video while keeping all other facial features and motion intact
Create comedic or stylized character effects without manual frame-by-frame editing

Workflow: IC-LoRA Video Input

Beta

IC-LoRAs in beta testing. These are functional but may have limitations or see breaking changes.

HDR (Beta)

Converts SDR video to HDR, producing per-frame EXR files for professional color grading. HDR is a post-processing tool that operates on existing video rather than guiding generation.

Model: LTX-2.3-22b-IC-LoRA-HDR (ltx-2.3-22b-ic-lora-hdr-0.9.safetensors)

Compatibility: LTX-2.3

For setup, usage, and DaVinci Resolve import instructions, see the HDR Video guide.

LipDub (Beta)

Re-generates speech in video, producing lip-synced output with new dialogue while preserving the speaker’s visual appearance and vocal identity. Works with live-action and animated subjects. LipDub is a video-to-video dubbing tool focused on speech replacement — it can also be used for rephrasing dialogue in the original language.

Model: LTX-2.3-22b-IC-LoRA-LipDub (ltx-2.3-22b-ic-lora-lipdub-0.9.safetensors)

Compatibility: LTX-2.3

For setup, usage, and prompting instructions, see the LipDub guide.

Resources

Official LTX IC-LoRAs on HuggingFace: