Edit Anything — Experimental LTX-2 Video Editing LoRAs

Heads up. These LoRAs are research experiments. They are far from production-ready and will fail on many inputs. They are released for the community to play with and break, not as a finished tool.

1. Edit Anything v2 for LTX-2.5 — the current one, a plain instruction-editing IC-LoRA. Start here. (built on LTX-2.5 22B)
2. Edit Anything v0.1 — motion transfer LoRA (two ranks). (LTX-2.3)
3. Edit Anything — no-reference multitask LoRA (rank 256, prompt-driven only). (LTX-2.3)
4. Reference video-to-video (Ref V2V) — experimental IC-LoRA + sidecar modules (two builds). (LTX-2.3)

0. Edit Anything v2 — LTX-2.5

What it is

Examples

edit_anything: Replace the green indoor background with a sandy beach under a bright sky, where the man
dances in dark blue swim trunks and a white beach towel draped over his shoulder, holding a
yellow inflatable ring around his waist.

edit_anything: Replace the purple Dodge Challenger in the background with a green Lamborghini Huracan,
featuring black five-spoke wheels and a low front splitter. The woman remains fully visible
throughout the video, wearing her brown suit as she dances in the foreground.

edit_anything: Add a cinematic apocalyptic scene with falling bombs and explosive detonations, creating
dramatic smoke and debris as the blue sports car races through the curves.

edit_anything: Replace the woman's dark black hair with vibrant red hair. Add a metallic leather
metalworker outfit, including a sleeveless top and pants, worn by her in the center of the
frame.

How to use it in ComfyUI

Prompting

Replace <the existing thing> with <the new thing, described>.
Add a <object with attributes> <where>.
Remove <the thing, identified precisely>.
Restyle <the region, identified> into <style> style.

• 10 to 20 words per instruction. Shorter loses the description the model needs; longer starts to drift.
• One object per Add. Write Add a ... / Add an .... For a group or a crowd use Replace on an existing region instead of Add several ....
• Anchor each instruction to something visible: "to the left of the man", "on the desk on the right side of the frame", "behind the speaker".
• Describe state, not manner. "standing in the green grass field", "sitting upright on the desk" — not "swimming gracefully".
• Restyle targets a region, not the whole clip. Name a wall, a screen, a background or a face. For a whole-video look, write two or three Restyle instructions over the regions that carry it.
• Make it <style> and Turn it into <style> do not work. Use Restyle.

edit_anything: Add a medium-sized brown dog sitting on the grass to the right of the
motorcycles. Remove the black jacket worn by the smiling man in the left foreground.
Replace the man's graying hair with bright red hair.

edit_anything: Replace the gray and black hoodie worn by the man with a bright red zip-up
jacket with white stripes on the sleeves.

edit_anything: Remove the smartphone held in the man's hand.

edit_anything: Restyle the large blue digital screen behind the man into flat vector style.

Training

Limitations

• The editing vocabulary is limited; requests far from the four verbs above get weaker results.
• Edits that change very little of the frame (subtle removals) are the least reliable.
• No reference image support — you cannot say "add this object" with a picture. That is what track 3 tried, and it is still unsolved.
• 704x384x73 is the trained box; longer clips lose temporal consistency first.

1. Edit Anything v0.1 (motion transfer)

• edit_anything_30k_v0.1_motion_transfer_r128.safetensors
• edit_anything_30k_v0.1_motion_transfer_r256.safetensors

What it is

1. Stage 1 — image-only pretraining. ~30 000 image edit pairs. Training a video model on still images is admittedly not ideal, but it was a way to push the editing vocabulary beyond what a small video-only dataset can teach.
2. Stage 2 — video fine-tune with first_frame_conditioning > 0. This restored the temporal prior and unlocked the motion-transfer behaviour described below.

Motion transfer

1. Take a guide video.
2. Replace its first frame with an edited still (insert a new subject, swap an object, etc.). Use a strong image-editing model — Flux Klein or similar — to prepare it; the quality of this single frame propagates through the whole clip.
3. Feed the edited frame as the first frame of the input, and the original guide video as the motion source.

• Hard scene cuts break it. The model assumes continuous motion from the first frame onwards. A cut to a different camera angle or location mid-clip will produce smearing, ghosting, or the inserted subject jumping to the wrong position. Use clips without cuts, or split at the cuts and process each segment separately.
• Very fast motion fails. Quick pans, fast subject movement, or high-velocity action confuse the motion-copy mechanism. Outputs degrade to blur or to the model "freezing" on the first-frame appearance and losing the motion entirely. Stick to moderate-speed clips.
• Poor blending / artefacts in the first frame propagate everywhere.
• Works best when the inserted subject roughly occupies the same region as whatever it replaces.

Prompting

Which rank to use

How to wire the LoopingSampler

• editanything_module → leave disconnected.
• ref_image → the edited first frame (for motion transfer) or the source frame you want preserved (for plain editing).
• guide_frames → the guide video.
• enable_role_embedding, enable_adaln, enable_visual_crossattn → all off. None of those branches were trained for v0.1; turning them on with no module connected does nothing anyway, but keeping them off silences the WARN logs.

2. Edit Anything — no-reference multitask LoRA

• edit_anything_v1.1_r256.safetensors

What it is

What it's different about it (vs v0.1)

• Two-stage training continuation: the first stage gave the model its edit vocabulary; the second stage refined it on a larger, more balanced video pair set covering Add / Remove / Replace / Style.
• Rank 256 (vs v0.1's effective rank from the merge), giving more capacity for the broader task mix.
• Trained directly on video pairs, so the temporal behaviour on these tasks tends to be steadier than on a model whose first stage was on still images.

How to use it

Limitations

• No reference image → identity is not anchored, so Add / Replace of a specific person or object will be wobblier than the Ref V2V build.
• No motion transfer (that's v0.1 only).

Prompting

Add — 15 to 30+ words, describe what to add and where

• Add a smiling woman with brown hair, wearing a pink sleeveless top, sitting to the right of the man at the news desk.
• Add a person wearing a blue denim shirt over a white t-shirt to the right side of the frame, behind the person cooking.
• Add a decorated Christmas tree with red and white ornaments and lights to the right of the man.
• Add a blonde boy wearing a black t-shirt with a blue collar and blue patterned pants, sitting behind the other children in the upper center of the frame.
• Add two horizontal wooden strips to the front of the white range hood.

Remove — very short, 4 to 10 words

• Remove the man drinking from a glass.
• Remove the disco ball.
• Remove the large tree on the right.
• Remove the squirrel in the foreground.
• Remove the man on the left.

Replace — 20 to 35 words, describe both old and new

• Replace the white panel door on the right side of the frame with a dark brown grandfather clock.
• Replace the light-colored cat lying on the mat on the floor with a young woman sitting on the mat.
• Replace the dark grey knitted sweater on the man's torso with a black and white patterned Christmas sweater.
• Replace the blue robot with a glowing blue face on the left with a smiling man wearing sunglasses and a blue shirt.
• Replace the sitting person wearing a black cape on the left with a black fabric draped over an object.

Style — fixed template, the style name is what changes

• Convert the video into a Pencil Sketch style.
• Convert the video into a Watercolor Painting style.
• Convert the video into a Van Gogh style.
• Convert the video into a Play-Doh style.
• Convert the video into a Claymation style.
• Convert the video into a 3D Chibi style.
• Convert the video into a Ghibli style.
• Convert the video into a Pop Art style.
• Convert the video into an American Cartoon style.
• Convert the video into a Flat Vector Cartoon style.

What it does not do

• No compositional prompts. "Add X and remove Y", "Replace A with B and add C", etc. are not in the training set. Captions combining two action verbs are essentially absent (the only ones present are the "Remove X and replace with Y" idiom, which is really a single Replace). Pure multi-action edits will fall apart — split them into separate runs.
• No "change background" as a task. Background is only used as a positional reference ("in the background", "on the wall in the background"). To swap the entire backdrop, phrase it as a Replace on a concrete background element, e.g. "Replace the brick wall in the background with a forest at sunset". Vague prompts like "Change the background to a beach" are off- distribution and rarely work.
• No global colour grade / lighting change. Only the Style template is trained as a global transform. Anything else global (LUT-style adjustments, time-of-day swaps without a concrete object) is unreliable.

Which LoRA should I use?

How to wire the LoopingSampler

• editanything_module → leave disconnected.
• ref_image → leave disconnected. This LoRA has no reference-image path; passing one will just pre-encode tokens that you do not want.
• guide_frames → the guide video.
• enable_role_embedding, enable_adaln, enable_visual_crossattn → all off (no module = nothing to inject anyway).

3. Reference video-to-video (Ref V2V) — experimental

• edit_anything_reference_v0.1_r128_ref_adaln_proj-role_embedding.standard.safetensors
• edit_anything_reference_v0.1_r128_ref_adaln_proj-role_embedding.module.safetensors
• edit_anything_reference_v0.1_r128_ref_adaln_proj-role_embedding-ref_attn-ref_visual_proj.standard.safetensors
• edit_anything_reference_v0.1_r128_ref_adaln_proj-role_embedding-ref_attn-ref_visual_proj.module.safetensors

What it is

Architecture — why this LoRA has "modules"

• ref_visual_proj — projects the reference VAE latent into 32 visual memory tokens.
• ref_attn — a dedicated cross-attention branch inside each transformer block, reading those tokens.
• ref_adaln_proj — a global AdaLN bias derived from the reference (palette / overall look).
• role_embedding — an experimental token bias inspired by some of Kijai's tests; whether it actually helps is still unclear.

How to load

1. Install the ref_attn cross-attention branch on every transformer block.
2. Inject the AdaLN / role / visual cross-attention conditioning at the correct points in the model.
3. Sample long videos in overlapping chunks with the conditioning re-applied per chunk.

How to wire the LoopingSampler

• Load the *.standard.safetensors through a normal ComfyUI LoraLoader before the sampler.
• Load the *.module.safetensors through LTXVEditAnythingModuleLoader and connect its editanything_module output to the sampler.
• On the sampler:
  • editanything_module → the module loader output (required).
  • ref_image → the reference image (required — this is what Add / Replace will insert).
  • guide_frames → the source video to edit.
  • enable_adaln → on (defaults match training).
  • enable_visual_crossattn → on for the 4-extras build; off (or will be a no-op) for the 2-extras build.
  • enable_role_embedding → off for the 4-extras build (training config disabled it). On if you're loading the 2-extras build alone.

Which build to use

• ref_adaln_proj-role_embedding — the original training. Only ships the two side-channel modules.
• ref_adaln_proj-role_embedding-ref_attn-ref_visual_proj — the continuation. Adds the visual cross-attention branch and its projector on top.

Reading the layers

• lora_layers_reference.md — full tensor inventory of both builds.
• lora_layers_impact.md — what each branch contributes at inference and which inference knob (adaln_scale, ref_context_scale, ref_token_scale, ref_start_block, ref_end_block, etc.) maps back to which training default.

Prompt examples

Edit Anything v0.1 — standard editing

• "Replace the stone statue of a man on the left with a young woman in a green dress."
• "Add a black labrador retriever sitting beside the woman on the bench."
• "Remove the teacher from the classroom."
• "Alter the cap's colour from modern black to deep maroon."
• "Replace the fresh citrus-green background with a wooden desk."
• "Add faint tire tracks across the snow behind the car."
• "Add a black statue, a blue camera, a cyan towel, a red guitar and a pink backpack to the lakeside pier."

• Imperative verbs: Add / Replace / Remove / Alter / Change.
• When replacing, describe both the original and the new subject so the model can localise the edit.
• Keep captions short and concrete. Long flowery prose hurts.

Edit Anything v0.1 — motion transfer

1. Pick a guide video.
2. Edit only the first frame externally (Flux Klein or any capable image-edit model) to introduce the new subject in the desired pose and position.
3. Feed the edited frame as the first frame of the input and the original guide as motion source.
4. The prompt should describe the inserted subject and the action being preserved.

• "Replace the standing man holding the umbrella with a woman in a red coat holding the same umbrella, walking across the puddles."
• "Add a tabby cat curled up in the armchair while the man in the background keeps reading."
• "Replace the runner in the blue jersey with a man wearing a white shirt and grey shorts running along the same path."

Reference V2V (Ref V2V) — Add and Replace

• "Add a middle-aged man with curly grey hair, a beard and glasses, wearing a blue quarter-zip sweater, on the right side of the frame, standing in front of a raw cut of meat on a tray."
• "Add a light-coloured small boat with dark seats and an outboard motor floating in the water."
• "Add an open book filled with colourful pencils in the woman's hands."
• "Add a silver metallic bucket on the table in front of the blonde character, with her hands stirring a mixture inside."
• "Add two miniature dolls, one blonde and one brunette, dressed in patterned clothing, sitting at a small table with teacups and small white vases on the countertop."

• "Replace the standing kangaroo holding the bicycle handlebars with a man wearing a white t-shirt, light brown shorts and a yellow cap, holding the bicycle handlebars."
• "Replace the stone statue of a man on the left side with a young woman in a green dress."
• "Replace the wooden barrel near the entrance with a large brown leather suitcase."

• Describe the inserted subject in full, even though the reference image is the source of truth — the text path drives placement and pose.
• For Replace, also describe what is being replaced so the model can match the spatial region.
• Keep the inserted subject roughly in the same scale and region as what it replaces.
• The captions in the training set average ~25–40 words — aim for that range. Single-sentence captions like "Add a man" are far too sparse and will fail.

Inference tips (applies to all models)

• If the model is ignoring the prompt (edit isn't being applied, the reference is barely showing up, the style transfer is faint), raising CFG is the single most common fix. Go up to 6–8 if needed.
• If results look over-saturated, plasticky, or motion is freezing, CFG is too high — pull back toward 3–4 or re-enable the distilled LoRA for CFG 1 if you don't actually need stronger prompt adherence.
• Ref V2V in particular benefits from being more aggressive with CFG when the reference identity isn't transferring cleanly.
• Combine CFG tuning with the LoRA-specific knobs from each section (adaln_scale, ref_context_scale, ref_token_scale for Ref V2V; prompt rewriting for v0.1 / no-ref).

ComfyUI nodes

• LTXVEditAnythingApply — load the LoRA + extras and patch the model.
• 🅛🅣🅧 LTXV Edit Anything Looping Sampler — sampler that injects role / AdaLN / visual cross-attention and handles long videos in chunks.
• LTXVEditAnythingModuleLoader — load the *.module.safetensors sidecar.

Status

Credits

Edit Anything LoRAs by Alisson Pereira dos Anjos (huggingface.co/Alissonerdx/EditAnything).