Add char_masks preprocessing script for SCAIL-2 training (task 3.7)

scripts/process_char_masks.py turns per-sample character label-map videos/images
(integer pixel labels: 0 = environment, 1..K = characters -> binding slots) into
the pixel-space semantic-mask tensors the SCAIL training/inference path consumes:
{"mask": [K+1, F_pix, H_pix, W_pix]} (ch0 = environment switch, ch1..K = slots).

- Aligns to the target video's latent grid read from the saved latent metadata
  (F_pix=(F-1)*8+1, H*32, W*32), so char_masks/ lines up file-for-file with
  latents/ / driving_latents/ for PrecomputedDataset.
- Nearest-neighbour resize so integer labels are never blended; labels > K are
  dropped with a warning; ch0 filled uniformly with --environment-switch.
- Reuses process_videos.py helpers (naming, atomic save, VAE factors) and matches
  its typer CLI conventions.

Verified on CPU: a synthetic 2-character label map (plus an out-of-range id)
produces mask (7,17,128,128) with ch0 uniform, slots placed correctly, id>K
dropped, and feeds encode_mask_channels to the 8*(K+1)=56 channels. README +
docs/tasks.md 3.7 updated (upstream label-map generation via SAM/tracking is
dataset-specific and still out of scope).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

docs/plan.md

@@ -0,0 +1,70 @@
+# SCAIL-2 → LTX-2 移植計畫
+
+將 SCAIL-2（arXiv:2606.10804，`zai-org/SCAIL-2`）的端到端角色動畫手法移植到 LTX-2。
+
+> **注意**：SCAIL-2 原始實作建構於 **Wan 2.1**，非 LTX-2。座標慣例與架構需翻譯到 LTX 的資料流。
+
+## 三個核心機制
+
+1. **Driving latent 直接串接** — 把驅動影片 latent 直接接進 DiT token 序列（不經骨架/pose 中介），用 width 軸座標偏移 ΔW 讓 driving 座標跟主 video 座標分開。
+2. **In-context mask channel** — 疊加額外輸入 channel（1 個環境開關 + K=6 個角色綁定槽，展開為 `4(K+1)=28` channel）到模型輸入，讓模型知道背景/角色對應。
+3. **Mode-specific RoPE** — Animation Mode 與 Replacement Mode 用不同的座標指派規則。
+
+## LTX-2 對應落點
+
+| 機制 | LTX-2 落點 | 改權重？ |
+|---|---|---|
+| 1. Driving 串接 + ΔW | 新 `ConditioningItem`（clone `reference_video_cond.py`），改 `positions` 偏移 | 否 |
+| 3. Mode-specific RoPE | 上述 item 加 `DrivingMode` enum | 否 |
+| 2. In-context mask channel | 加寬 `patchify_proj` 輸入 channel + `LatentState` 帶額外 channel + 投影前 concat | **是**（需微調） |
+
+**關鍵洞察**：LTX 的 `ConditioningItem.apply_to(latent_state, latent_tools) -> LatentState` 就是「串接進序列」的天然注入點，機制 1+3 **完全不用動 transformer / rope.py**。RoPE 由 `LatentState.positions` `[B,3,T,2]`（pixel 座標，axis1=(time,h,w)）驅動；ΔW = 對 `positions[:,2]`（width）加常數。width 正規化上界 `max_pos[2]=2048`，超過會 wrap。
+
+## SCAIL-2 座標規則（論文）
+
+序列 `[z_ref; z_t; z_driv]`，driving 永遠在 width 軸帶固定偏移 ΔW：
+
+| | Animation | Replacement |
+|---|---|---|
+| z_ref | T=0, H=[0,Hv), W=[0,Wv) | T=0, **H=[ΔH_ref, ΔH_ref+Hv)**, W=[0,Wv) |
+| z_t | T=[1,Tv], H=[0,Hv), W=[0,Wv) | T=[0,Tv−1], H=[0,Hv), W=[0,Wv) |
+| z_driv | T=[1,Tv], H=[0,Hv), **W=[ΔW, ΔW+Wv)** | T=[0,Tv−1], H=[0,Hv), **W=[ΔW, ΔW+Wv)** |
+
+## 分階段計畫
+
+### Phase 1 — Driving 串接 item（機制 1+3，推論期 PoC）✅ 已完成
+- 純推論、不改權重、不動 `patchify_proj`、不碰 ltx-trainer。
+- 用現有 checkpoint 驗證資料流正確性（序列長度、座標偏移、attention mask、denoise_mask、不 wrap）。
+- **PoC 僅驗證 plumbing**；LTX-2 未經此訓練，畫面不會是正確動畫。
+
+### Phase 2 — In-context mask channel（機制 2，checkpoint 手術）✅ 已完成（plumbing）
+- **決策**：時間編碼採忠實堆疊，`8×(K+1)=56` channel（LTX 時間因子 8，K=6），非 Wan 的 28。
+- `LTXModel(mask_conditioning_channels=0)` config-gated 加寬 `patchify_proj`；預設不變。
+- `LatentState`/`Modality` 加 `cond_channels` 欄位，跟著 clone/clear/append 流動；投影前在 `_apply_patchify_proj` concat（None 補零）。
+- 新輸入欄位 **zero-init**：`widen_patchify_proj_for_mask_channels` 轉換舊 checkpoint，行為不變、待微調才生效。
+- `VideoConditionByMaskChannels`：(K+1) 語意 mask → 空間下採樣 + 時間 8× 堆疊 → 寫入尾端 driving token（target 保持零，符合論文）。
+- 驗證通過（`verify_mask_channels.py`）：向後相容、zero-init 加寬 forward == baseline、mask pipeline 不 crash。**僅驗證 plumbing，畫質需 Phase 3 微調。**
+
+### Phase 3 — 訓練整合（ltx-trainer）✅ 已完成（程式碼路徑；實訓需 GPU）
+- **決策**：完整整合到 `FlexibleStrategy`；訓練 = LoRA + 解凍 `patchify_proj`（新 mask 欄位無法純 LoRA 訓練）。
+- 新 `DrivingConditionConfig` + `MaskChannelsConditionConfig`（`flexible.py`）；`_apply_driving_condition`(cond-first concat + ΔW) + `_build_mask_channels`(重用 `encode_mask_channels`) → `Modality.cond_channels`。
+- `load_transformer(mask_conditioning_channels=)` 用 `widen_module_patchify_proj_for_mask_channels` 加寬；`ModelConfig.mask_conditioning_channels`；trainer 在 LoRA 模式解凍 patchify_proj。
+- `configs/scail_animation_lora.yaml` + docs。CPU 單元驗證通過（`verify_phase3_trainer.py`）。
+- **本機無 GPU/Linux/checkpoint → 未跑實機訓練**；dataset 前處理（driving latents + 語意 mask）與 validation runner 接線未做。
+
+### Phase 4 — Pipeline + CLI 包裝 ✅ 已完成（程式碼路徑；實跑需 GPU）
+- `LTXModelConfigurator` 讀 `mask_conditioning_channels` → SCAIL checkpoint 自描述、載入自動加寬（不需 runtime wrapper）。
+- `scail_animation.py`：`ScailAnimationPipeline`（仿 `distilled.py` 兩階段）+ 可測純函式 `build_scail_conditionings`（driving + mask 條件組裝）+ `load_masks` + `main()`。
+- `utils/args.py` `scail_animation_arg_parser`（`--driving-video/--mask-path/--mode/--driving-strength`）。
+- CPU 驗證通過（`verify_phase4_pipeline.py`）；`ltx-pipelines/CLAUDE.md` 表格 row。
+- **實機端到端需 GPU + SCAIL-trained checkpoint**（config 含 `mask_conditioning_channels` + 訓練好的 patchify_proj + LoRA）。
+
+## Phase 1 簡化取捨（記錄，Phase 2 需回頭處理）
+- (a) driving 時間座標直接複製 target 的（token-wise），故 driving 需與 target 同 F/H/W。
+- (b) 單一 frozen driving group 的 `attention_mask` 維持 None（= 全連接，target 完全看得到 driving），與 reference cond 一致。
+- (c) **ANIMATION 與 REPLACEMENT 在 Phase 1 產生相同 driving 座標** — mode 差異（z_ref 的 ΔH_ref 高度位移、target 時間原點、mask channel）屬 Phase 2，enum 先保留佔位。
+
+## 參考
+- 論文：arXiv:2606.10804 — *SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning*
+- 官方實作：`zai-org/SCAIL-2`（GitHub / HuggingFace），建構於 Wan 2.1
+- 藍本檔案：`packages/ltx-core/src/ltx_core/conditioning/types/reference_video_cond.py`

docs/tasks.md

@@ -0,0 +1,59 @@
+# SCAIL-2 → LTX-2 移植任務追蹤
+
+狀態圖例：✅ 完成 ｜ 🔄 進行中 ｜ ⬜ 未開始 ｜ ⏸️ 暫緩
+
+相關計畫見 [`plan.md`](./plan.md)。
+
+## Phase 1 — Driving 串接 item（推論期 PoC）
+
+| # | 任務 | 狀態 | 產出 / 備註 |
+|---|---|---|---|
+| 1.1 | 新增 `VideoConditionByDrivingLatent` + `DrivingMode` | ✅ | `packages/ltx-core/src/ltx_core/conditioning/types/driving_video_cond.py`，以 `reference_video_cond.py` 為藍本，實作 ΔW width 偏移、時間對齊 target、`max_pos` 防呆、token 數防呆 |
+| 1.2 | 匯出新 conditioning 類別 | ✅ | `conditioning/types/__init__.py`、`conditioning/__init__.py` |
+| 1.3 | 免-GPU 資料流驗證腳本 | ✅ | 序列長度、ΔW 不重疊、時間對齊、frozen denoise_mask、attention_mask=None、`clear_conditioning` 剝除、超界/token 數防呆 — 全通過；ruff clean |
+| 1.4 | （選配）端到端 smoke run | ✅ | 本機 CPU-only、無 checkpoint，改用小型真實 `LTXModel`（隨機權重）跑 pipeline 實走路徑：`create_noised_state`（含 driving cond）→ `modality_from_latent_state` → **真 transformer forward**（seq 160=target 80+driving 80）→ `clear_conditioning`（→80）。ANIMATION/REPLACEMENT 皆通過：不 crash、輸出 finite、driving frozen 且正確剝除。**僅驗證整合，不評估畫質** |
+
+## Phase 2 — In-context mask channel（checkpoint 手術）
+
+> **決策**：mask channel 時間編碼採**忠實堆疊**——每 latent 幀對應 8 個 pixel 子幀沿 channel 堆疊，`8×(K+1)=56` channel（LTX 時間因子 8，K=6）。非 Wan 的 4×(K+1)=28。
+
+| # | 任務 | 狀態 | 備註 |
+|---|---|---|---|
+| 2.1 | `patchify_proj` 加寬（config-gated） | ✅ | `LTXModel(mask_conditioning_channels=0)`；>0 時 `patchify_proj=Linear(in+mask, inner)`。預設不變 |
+| 2.2 | `LatentState`/`Modality` 帶 `cond_channels` 欄位 | ✅ | `types.py`、`modality.py` 加 `cond_channels: Tensor\|None=None`（patchified [B,T,C]）；`tools.clear_conditioning` 裁切；`helpers.modality_from_latent_state` 帶入 |
+| 2.3 | 投影前 concat/zero-pad cond_channels | ✅ | `transformer_args.py` `_apply_patchify_proj`：寬度不足時用 cond_channels 補齊，None 則補零 |
+| 2.4 | checkpoint zero-init 加寬轉換 | ✅ | `mask_channels_checkpoint.py` `widen_patchify_proj_for_mask_channels`：尾端補零欄，載入舊權重行為不變 |
+| 2.5 | `VideoConditionByMaskChannels`（56ch） | ✅ | `mask_channels_cond.py`：1 環境開關 + K=6 綁定槽 → 空間下採樣 + 時間 8× 堆疊(causal 首幀複製)=56ch，寫入尾端 driving token，target 保持零 |
+| 2.6 | token-append conditioning 延伸 cond_channels | ✅ | `cond_channels.py` `extend_cond_channels`；reference_video/reference_audio/driving/keyframe 皆接上 |
+| 2.7 | Phase 2 驗證（免訓練） | ✅ | `verify_mask_channels.py`：mask=0 向後相容；zero-init 加寬 forward == baseline（任意 cond_channels）；driving+mask pipeline forward 不 crash、cond_channels 形狀/placement 正確、`clear_conditioning` 剝除 |
+| 2.8 | Replacement Mode z_ref 高度位移 ΔH_ref | ⬜ | 仍暫緩（Phase 1 divergence，需獨立 reference token group） |
+
+## Phase 3 — 訓練整合（ltx-trainer）
+
+> **決策**：完整整合到 `FlexibleStrategy`；訓練方式 = **LoRA + 解凍 patchify_proj**（新 mask 欄位無法純 LoRA 訓練）。**本機無 GPU/Linux/checkpoint，只做到 CPU 單元驗證**，實訓需在 GPU 機器跑。
+
+| # | 任務 | 狀態 | 備註 |
+|---|---|---|---|
+| 3.1 | 新增 Driving/MaskChannels ConditionConfig | ✅ | `flexible.py`：`DrivingConditionConfig`(latents_dir/mode/width_offset) + `MaskChannelsConditionConfig`(mask_dir/num_slots)，加入 union + `get_data_sources` |
+| 3.2 | strategy 接線 driving concat + cond_channels | ✅ | `_apply_driving_condition`(cond-first concat + ΔW width 偏移) + `_build_mask_channels`(重用 `encode_mask_channels`，寫前 N driving token) → `Modality.cond_channels` |
+| 3.3 | model_loader + ModelConfig 支援加寬 | ✅ | `widen_module_patchify_proj_for_mask_channels`(live module zero-init) + `load_transformer(mask_conditioning_channels=)` + `ModelConfig.mask_conditioning_channels` |
+| 3.4 | LoRA 模式解凍 patchify_proj | ✅ | `trainer._unfreeze_patchify_proj`：mask_channels>0 時把 video patchify_proj 設 trainable，讓新欄位隨 LoRA 一起訓 |
+| 3.5 | 範例 config + docs | ✅ | `configs/scail_animation_lora.yaml`；`configs/README.md` 與 `docs/training-modes.md` 表格 row |
+| 3.6 | CPU 單元驗證 | ✅ | `verify_phase3_trainer.py`：config round-trip、prepare_training_inputs 建 cond_channels[B,T,56]、driving 前置/mask placement、widened model forward + compute_loss finite、widen helper zero-init 等價 |
+| 3.7 | dataset 前處理（產 driving latents + 語意 mask） | 🟡 | `char_masks/` 前處理已做：`scripts/process_char_masks.py`（label-map 影片/圖 → `[K+1,F_pix,H,W]`，對齊 target latent，nearest 保留整數 label，ch0 環境開關）。`driving_latents/` 沿用既有 `process_videos.py`（driving 影片走 video latent 路徑，同 target 形狀）。**仍缺**：從原始影片產生 label-map 的分割/追蹤步驟（SAM 等，資料集特定，未含） |
+| 3.8 | validation runner 接 driving/mask | ⬜ | 驗證期取樣尚未接 SCAIL 條件（config 內 validation 先停用），與 Phase 4 一起 |
+| 3.9 | 實機訓練跑通 | ⬜ | 需 Linux + GPU + checkpoint，本機無法 |
+
+## Phase 4 — Pipeline + CLI
+
+| # | 任務 | 狀態 | 備註 |
+|---|---|---|---|
+| 4.1 | configurator 讀 `mask_conditioning_channels` | ✅ | `LTXModelConfigurator` 兩分支 `config.get("mask_conditioning_channels",0)` → SCAIL checkpoint 自描述、載入自動加寬（不需 runtime widen wrapper） |
+| 4.2 | `ScailAnimationPipeline` + `build_scail_conditionings` | ✅ | `scail_animation.py`：仿 `distilled.py` 兩階段，注入 SCAIL driving+mask conditioning；`build_scail_conditionings` 為可測純函式；`load_masks` helper |
+| 4.3 | `scail_animation_arg_parser` + `main()` | ✅ | `utils/args.py`：在 `default_2_stage_distilled_arg_parser` 上加 `--driving-video/--mask-path/--mode/--driving-strength` |
+| 4.4 | CPU 驗證 + docs | ✅ | `verify_phase4_pipeline.py`：import、CLI round-trip、conditioning assembly(driving append + mask 在尾端)、configurator 自描述加寬；`ltx-pipelines/CLAUDE.md` pipeline 表格 row |
+| 4.5 | 實機端到端跑通 | ⬜ | 需 GPU + SCAIL-trained checkpoint（含 `mask_conditioning_channels` config + 訓練好的 patchify_proj + LoRA） |
+
+## 決議紀錄
+- **範圍**：先只做 Phase 1（推論期 PoC）。Phase 2+ 待 Phase 1 驗證後再討論。
+- **架構前提**：SCAIL-2 建構於 Wan 2.1，本移植為跨架構移植。

modal/README.md

@@ -0,0 +1,111 @@
+# SCAIL-2 on Modal — free testing environment
+
+Test the SCAIL-2 LTX-2 training/inference **code path** on Modal without a local
+GPU. The free path uses tiny random-init models + synthetic data (no checkpoint,
+no dataset), so it validates that the SCAIL wiring runs on real Linux/GPU — not
+model quality.
+
+> ⚠️ The real 19B LTX-2 model is **not** free to train/run. The `verify`/`smoke`
+> targets are near-free; `train` on a real checkpoint uses a paid GPU.
+
+## 1. One-time setup
+
+1. Sign up at [modal.com](https://modal.com) (the free Starter plan includes a
+   monthly credit allowance — enough for many `verify`/`smoke` runs).
+2. Install and authenticate:
+   ```bash
+   pip install modal
+   modal setup          # opens a browser to link your account / token
+   ```
+
+## 2. Free / near-free checks
+
+From the repo root:
+
+```bash
+# CPU: SCAIL-2 Phase 1-4 plumbing (driving concat, mask channels, zero-init widen,
+# a FlexibleStrategy training step + loss, inference conditioning assembly).
+modal run modal/app.py::verify
+
+# Same checks on a T4 GPU (validates the CUDA path). ~cents.
+modal run modal/app.py::smoke
+```
+
+The first run builds the image (installs the `ltx-core`/`ltx-pipelines`/`ltx-trainer`
+workspace via `uv sync`). `ltx-kernels` (CUDA-compiled) is intentionally skipped —
+attention falls back to PyTorch SDPA, so no CUDA toolchain is required.
+
+Expected tail:
+```
+[OK] Phase 1 driving concat: seq 48 -> 96
+[OK] Phase 2 zero-init widen: in_features=72, output preserved
+[OK] Phase 3 training step: cond_channels (1, 96, 56), loss ...
+[OK] Phase 4 build_scail_conditionings: [driving, mask_channels]
+All SCAIL-2 checks passed on cuda   # (or cpu)
+```
+
+## 3. Scaling up to real training (paid)
+
+`train` runs `packages/ltx-trainer/scripts/train.py` against a real checkpoint.
+You must supply the weights + data via the persistent `scail-data` Volume.
+
+1. Create/populate the Volume (checkpoint, Gemma encoder, preprocessed latents):
+   ```bash
+   modal volume create scail-data          # if not auto-created
+   modal volume put scail-data /local/ltx-2-model.safetensors /model/ltx-2.safetensors
+   modal volume put scail-data /local/gemma /model/gemma
+   modal volume put scail-data /local/preprocessed /data/preprocessed
+   ```
+2. Copy `configs/scail_animation_lora.yaml`, and point its paths at the mounted
+   Volume (everything lands under `/data` in the container):
+   ```yaml
+   model:
+     model_path: "/data/model/ltx-2.safetensors"
+     text_encoder_path: "/data/model/gemma"
+     mask_conditioning_channels: 56       # widens patchify_proj at load
+   data:
+     preprocessed_data_root: "/data/preprocessed"
+   ```
+   (Put your edited config on the Volume too, or bake it into the repo.)
+3. Launch (pick a GPU big enough for the model — the 19B needs an A100):
+   ```bash
+   # edit gpu="A10G" -> "A100" in modal/app.py::train for the full model
+   modal run modal/app.py::train --config-rel /data/scail_animation_lora.yaml
+   ```
+
+### Dataset preprocessing (not yet automated for SCAIL)
+
+The SCAIL training config expects, under `preprocessed_data_root/`:
+
+```
+latents/           # target video latents
+conditions/        # text embeddings
+driving_latents/   # driving video latents (same F/H/W as target)
+char_masks/        # per-sample "mask" = [K+1, F_pix, H, W]  (ch0 env switch, 1..K binding slots)
+```
+
+`latents/`, `conditions/`, and `driving_latents/` come from the existing
+`packages/ltx-trainer/scripts/process_dataset.py` (run it once per video set).
+`char_masks/` is produced by `packages/ltx-trainer/scripts/process_char_masks.py`
+from per-sample **label-map** videos/images (integer pixel labels: `0` =
+environment, `1..K` = characters → binding slots):
+
+```bash
+python packages/ltx-trainer/scripts/process_char_masks.py dataset.csv \
+    --mask-column char_labels --latents-dir ./latents \
+    --output-dir ./char_masks --num-slots 6 --main-media-column media_path
+```
+
+You still need a segmentation/tracking model (e.g. SAM) to *produce* those label
+maps from raw video — that upstream step is dataset-specific and not included.
+
+## Cost notes
+
+| Target  | GPU   | Rough cost | Use |
+|---------|-------|-----------|-----|
+| `verify`| none  | ~free      | validate code path on CPU |
+| `smoke` | T4    | cents      | validate CUDA path |
+| `train` | A10G/A100 | paid   | real fine-tuning (needs checkpoint + data) |
+
+Free credits are best spent on `verify`/`smoke` to catch integration issues before
+committing a paid GPU to a real run. Watch usage in the Modal dashboard.

modal/app.py

@@ -0,0 +1,84 @@
+"""Modal app for testing the SCAIL-2 LTX-2 training/inference path.
+
+Free-tier friendly: `verify` runs on CPU and `smoke` on a cheap T4, both using
+tiny random-init models + synthetic data (no checkpoint, no dataset) so a full run
+costs pennies. `train` is the scale-up entrypoint that runs the real trainer once
+you upload a checkpoint + preprocessed data to the `scail-data` Volume -- that one
+uses a real GPU and is NOT free.
+
+Setup (once):
+    pip install modal && modal setup
+
+Run:
+    modal run modal/app.py::verify     # CPU plumbing check (~free)
+    modal run modal/app.py::smoke      # same checks on a T4 GPU (cents)
+    modal run modal/app.py::train --config-rel configs/scail_animation_lora.yaml
+"""
+
+import subprocess
+from pathlib import Path
+
+import modal
+
+REPO = Path(__file__).parent.parent
+
+# Build the workspace once into the image. ltx-kernels (CUDA-compiled) is excluded
+# from the workspace and not needed -- attention falls back to SDPA.
+image = (
+    modal.Image.debian_slim(python_version="3.11")
+    .apt_install("git")
+    .pip_install("uv")
+    .env({"UV_LINK_MODE": "copy", "UV_PROJECT_ENVIRONMENT": "/root/LTX-2/.venv"})
+    .add_local_dir(
+        str(REPO),
+        "/root/LTX-2",
+        copy=True,
+        ignore=[".git", ".venv", "**/__pycache__", "**/*.pyc", "outputs", "wandb", "**/.pytest_cache"],
+    )
+    .run_commands("cd /root/LTX-2 && uv sync --package ltx-trainer")
+)
+
+app = modal.App("scail-ltx2", image=image)
+
+# Persistent volume for the (large) checkpoint + preprocessed dataset used by `train`.
+data_volume = modal.Volume.from_name("scail-data", create_if_missing=True)
+
+_UV_PY = ["uv", "run", "--package", "ltx-trainer", "python"]
+
+
+def _run(args: list[str]) -> None:
+    subprocess.run([*_UV_PY, *args], cwd="/root/LTX-2", check=True)
+
+
+@app.function(timeout=1800)
+def verify() -> None:
+    """Run the SCAIL-2 plumbing checks on CPU (near-free)."""
+    _run(["modal/checks.py"])
+
+
+@app.function(gpu="T4", timeout=1800)
+def smoke() -> None:
+    """Run the same checks on a T4 GPU to validate the CUDA path (cents)."""
+    _run(["modal/checks.py"])
+
+
+@app.function(gpu="A10G", timeout=60 * 60 * 6, volumes={"/data": data_volume})
+def train(config_rel: str = "configs/scail_animation_lora.yaml") -> None:
+    """Run the real SCAIL trainer. NOT free -- needs a real checkpoint + data.
+
+    Upload your base checkpoint, Gemma text encoder, and preprocessed data to the
+    ``scail-data`` Volume (mounted at /data), and point the config's paths at /data.
+    For the full 19B model use a bigger GPU (e.g. gpu="A100") and expect real cost.
+    """
+    _run(["packages/ltx-trainer/scripts/train.py", config_rel])
+    data_volume.commit()
+
+
+@app.local_entrypoint()
+def main(target: str = "verify", config_rel: str = "configs/scail_animation_lora.yaml") -> None:
+    if target == "smoke":
+        smoke.remote()
+    elif target == "train":
+        train.remote(config_rel)
+    else:
+        verify.remote()

modal/checks.py

@@ -0,0 +1,159 @@
+# ruff: noqa: T201
+"""Device-aware SCAIL-2 smoke checks (CPU or GPU), runnable anywhere.
+
+Consolidates the Phase 1-4 plumbing checks into one script that auto-selects CUDA
+when available. It uses tiny, randomly-initialised models and synthetic data, so it
+needs **no checkpoint and no dataset** -- ideal for a near-free Modal test that
+validates the SCAIL training/inference code path end to end in a real Linux/GPU
+environment before spending credits on the full 19B model.
+
+Run locally:   uv run --package ltx-trainer python modal/checks.py
+On Modal:      modal run modal/app.py::verify        (CPU)
+               modal run modal/app.py::smoke         (T4 GPU)
+"""
+
+from __future__ import annotations
+
+from dataclasses import replace
+
+import torch
+
+from ltx_core.components.noisers import GaussianNoiser
+from ltx_core.components.patchifiers import VideoLatentPatchifier
+from ltx_core.conditioning import DrivingMode, VideoConditionByDrivingLatent
+from ltx_core.model.transformer.mask_channels_checkpoint import (
+    widen_module_patchify_proj_for_mask_channels,
+)
+from ltx_core.model.transformer.model import LTXModel, LTXModelType
+from ltx_core.tools import VideoLatentTools
+from ltx_core.types import SpatioTemporalScaleFactors, VideoLatentShape
+from ltx_pipelines.scail_animation import build_scail_conditionings
+from ltx_pipelines.utils.helpers import create_noised_state, modality_from_latent_state
+from ltx_trainer.timestep_samplers import UniformTimestepSampler
+from ltx_trainer.training_strategies.flexible import (
+    DrivingConditionConfig,
+    FlexibleStrategy,
+    FlexibleStrategyConfig,
+    MaskChannelsConditionConfig,
+    ModalityConfig,
+)
+
+# --- config ---------------------------------------------------------------------
+B, C, F, H, W = 1, 16, 3, 4, 4
+HEADS, HEAD_DIM, LAYERS, INNER = 4, 16, 2, 64
+CTX_LEN, K = 8, 6
+SCALE = SpatioTemporalScaleFactors.default()
+MASK_CH = SCALE.time * (K + 1)  # 56
+F_PIX = (F - 1) * SCALE.time + 1
+N = F * H * W
+
+
+def _tiny_model(mask_channels: int, device: torch.device, dtype: torch.dtype) -> LTXModel:
+    model = LTXModel(
+        model_type=LTXModelType.VideoOnly,
+        num_attention_heads=HEADS,
+        attention_head_dim=HEAD_DIM,
+        in_channels=C,
+        out_channels=C,
+        num_layers=LAYERS,
+        cross_attention_dim=INNER,
+        mask_conditioning_channels=mask_channels,
+    )
+    for p in model.parameters():
+        torch.nn.init.normal_(p, std=0.02)
+    return model.to(device=device, dtype=dtype).eval()
+
+
+def main() -> None:
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    dtype = torch.float32
+    print(f"== SCAIL-2 checks on device={device} "
+          f"({torch.cuda.get_device_name(0) if device.type == 'cuda' else 'cpu'}) ==")
+    torch.manual_seed(0)
+
+    tools = VideoLatentTools(
+        patchifier=VideoLatentPatchifier(patch_size=1),
+        target_shape=VideoLatentShape(batch=B, channels=C, frames=F, height=H, width=W),
+        fps=24.0,
+        scale_factors=SCALE,
+    )
+
+    # Phase 1: driving-latent concat + ΔW RoPE (inference conditioning path).
+    driving = torch.randn(B, C, F, H, W, device=device, dtype=dtype)
+    noiser = GaussianNoiser(generator=torch.Generator(device=device).manual_seed(1))
+    state = create_noised_state(
+        tools, [VideoConditionByDrivingLatent(driving, mode=DrivingMode.ANIMATION)], noiser, dtype, device, 1.0
+    )
+    assert state.latent.shape[1] == 2 * N
+    print(f"[OK] Phase 1 driving concat: seq {N} -> {state.latent.shape[1]}")
+
+    # Phase 2: widen patchify_proj (zero-init) is output-preserving.
+    base = _tiny_model(0, device, dtype)
+    ctx = torch.randn(B, CTX_LEN, INNER, device=device, dtype=dtype)
+    sigma = torch.ones(B, device=device, dtype=dtype)
+    mod = modality_from_latent_state(state, context=ctx, sigma=sigma)
+    with torch.inference_mode():
+        out0, _ = base(video=mod, audio=None, perturbations=None)
+    widen_module_patchify_proj_for_mask_channels(base, MASK_CH)
+    mod_cc = replace(mod, cond_channels=torch.randn(B, 2 * N, MASK_CH, device=device, dtype=dtype))
+    with torch.inference_mode():
+        out1, _ = base(video=mod_cc, audio=None, perturbations=None)
+    assert torch.allclose(out0, out1, atol=1e-4), "zero-init widening changed output"
+    print(f"[OK] Phase 2 zero-init widen: in_features={base.patchify_proj.in_features}, output preserved")
+
+    # Phase 3: FlexibleStrategy driving + mask-channels training step.
+    cfg = FlexibleStrategyConfig(
+        name="flexible",
+        video=ModalityConfig(
+            is_generated=True,
+            latents_dir="video_latents",
+            conditions=[
+                DrivingConditionConfig(type="driving", latents_dir="driving_latents"),
+                MaskChannelsConditionConfig(type="mask_channels", mask_dir="char_masks", num_slots=K),
+            ],
+        ),
+    )
+    strategy = FlexibleStrategy(cfg)
+
+    def latents() -> dict:
+        return {
+            "latents": torch.randn(B, C, F, H, W, device=device),
+            "num_frames": torch.tensor([F]),
+            "height": torch.tensor([H]),
+            "width": torch.tensor([W]),
+            "fps": torch.tensor([24.0]),
+        }
+
+    mask_pix = torch.rand(B, K + 1, F_PIX, H * SCALE.height, W * SCALE.width, device=device)
+    batch = {
+        "video_latents": latents(),
+        "driving_latents": latents(),
+        "char_masks": {"mask": (mask_pix > 0.5).float()},
+        "conditions": {
+            "video_prompt_embeds": torch.randn(B, CTX_LEN, INNER, device=device),
+            "audio_prompt_embeds": torch.randn(B, CTX_LEN, INNER, device=device),
+            "prompt_attention_mask": torch.ones(B, CTX_LEN, device=device),
+        },
+    }
+    inputs = strategy.prepare_training_inputs(batch, UniformTimestepSampler(0.0, 1.0))
+    assert inputs.video.cond_channels.shape == (B, 2 * N, MASK_CH)
+    model = _tiny_model(MASK_CH, device, dtype)
+    with torch.inference_mode():
+        vpred, _ = model(video=inputs.video, audio=None, perturbations=None)
+    loss = strategy.compute_loss(vpred, None, inputs)
+    assert loss.shape == (B,)
+    assert torch.isfinite(loss).all()
+    cc_shape = tuple(inputs.video.cond_channels.shape)
+    print(f"[OK] Phase 3 training step: cond_channels {cc_shape}, loss {loss.item():.4f}")
+
+    # Phase 4: inference conditioning assembly.
+    masks = batch["char_masks"]["mask"]
+    conds = build_scail_conditionings(driving, masks, mode=DrivingMode.ANIMATION)
+    assert len(conds) == 2
+    print("[OK] Phase 4 build_scail_conditionings: [driving, mask_channels]")
+
+    print("\nAll SCAIL-2 checks passed on", device)
+
+
+if __name__ == "__main__":
+    main()

packages/ltx-core/src/ltx_core/conditioning/__init__.py

@@ -5,10 +5,14 @@ from ltx_core.conditioning.item import ConditioningItem
 from ltx_core.conditioning.types import (
     AudioConditionByReferenceLatent,
     ConditioningItemAttentionStrengthWrapper,
+    DrivingMode,
+    VideoConditionByDrivingLatent,
     VideoConditionByKeyframeIndex,
     VideoConditionByLatentIndex,
     VideoConditionByMask,
+    VideoConditionByMaskChannels,
     VideoConditionByReferenceLatent,
+    encode_mask_channels,
 )
 
 __all__ = [
@@ -16,8 +20,12 @@ __all__ = [
     "ConditioningError",
     "ConditioningItem",
     "ConditioningItemAttentionStrengthWrapper",
+    "DrivingMode",
+    "VideoConditionByDrivingLatent",
     "VideoConditionByKeyframeIndex",
     "VideoConditionByLatentIndex",
     "VideoConditionByMask",
+    "VideoConditionByMaskChannels",
     "VideoConditionByReferenceLatent",
+    "encode_mask_channels",
 ]

packages/ltx-core/src/ltx_core/conditioning/cond_channels.py

@@ -0,0 +1,28 @@
+"""Helpers for in-context conditioning channels (``LatentState.cond_channels``).
+
+Conditioning channels are extra per-token input features (e.g. SCAIL-2 mask
+channels) that ride alongside the latent in patchified token space ``(B, T, C)``
+and are concatenated onto the latent before the model's first projection. When
+present they must stay length-aligned with the token sequence, so any
+conditioning item that *appends* tokens must also extend the channels.
+"""
+
+from __future__ import annotations
+
+import torch
+
+from ltx_core.types import LatentState
+
+
+def extend_cond_channels(latent_state: LatentState, num_new_tokens: int) -> torch.Tensor | None:
+    """Return ``cond_channels`` extended with ``num_new_tokens`` zero rows, or ``None``.
+
+    Appended tokens (reference / driving / keyframe) carry no in-context signal by default, so they
+    are padded with zeros to preserve the ``T``-alignment invariant. Returns ``None`` unchanged when
+    the state has no conditioning channels (the standard case), keeping non-SCAIL pipelines untouched.
+    """
+    cond_channels = latent_state.cond_channels
+    if cond_channels is None:
+        return None
+    zeros = cond_channels.new_zeros(cond_channels.shape[0], num_new_tokens, cond_channels.shape[2])
+    return torch.cat([cond_channels, zeros], dim=1)

packages/ltx-core/src/ltx_core/conditioning/types/__init__.py

@@ -1,8 +1,10 @@
 """Conditioning type implementations."""
 
 from ltx_core.conditioning.types.attention_strength_wrapper import ConditioningItemAttentionStrengthWrapper
+from ltx_core.conditioning.types.driving_video_cond import DrivingMode, VideoConditionByDrivingLatent
 from ltx_core.conditioning.types.keyframe_cond import VideoConditionByKeyframeIndex
 from ltx_core.conditioning.types.latent_cond import VideoConditionByLatentIndex
+from ltx_core.conditioning.types.mask_channels_cond import VideoConditionByMaskChannels, encode_mask_channels
 from ltx_core.conditioning.types.mask_cond import VideoConditionByMask
 from ltx_core.conditioning.types.reference_audio_cond import AudioConditionByReferenceLatent
 from ltx_core.conditioning.types.reference_video_cond import VideoConditionByReferenceLatent
@@ -10,8 +12,12 @@ from ltx_core.conditioning.types.reference_video_cond import VideoConditionByRef
 __all__ = [
     "AudioConditionByReferenceLatent",
     "ConditioningItemAttentionStrengthWrapper",
+    "DrivingMode",
+    "VideoConditionByDrivingLatent",
     "VideoConditionByKeyframeIndex",
     "VideoConditionByLatentIndex",
     "VideoConditionByMask",
+    "VideoConditionByMaskChannels",
     "VideoConditionByReferenceLatent",
+    "encode_mask_channels",
 ]

packages/ltx-core/src/ltx_core/conditioning/types/driving_video_cond.py

@@ -0,0 +1,162 @@
+"""Driving-video conditioning for SCAIL-2-style end-to-end character animation.
+
+Ports mechanism 1 + 3 of SCAIL-2 (arXiv:2606.10804) to LTX-2: the *driving*
+video latent is concatenated directly into the DiT token sequence (no skeleton /
+pose intermediate), and carries a fixed spatial offset ``width_offset`` (the
+paper's ΔW) on the RoPE width axis so its coordinates stay detached from the
+main video tokens. This is the inference-only PoC path -- it reuses the existing
+frozen-reference-token machinery and does not touch the transformer or RoPE.
+
+This mirrors :class:`ltx_core.conditioning.types.reference_video_cond.VideoConditionByReferenceLatent`
+(same patchify -> positions -> append -> attention-mask flow); the only new
+behaviour is the mode-aware coordinate assignment.
+
+Scope note (Phase 1): SCAIL-2's in-context mask channels (mechanism 2) and the
+reference-latent height shift ΔH_ref of Replacement Mode require model-weight
+surgery / a separate reference token group and are intentionally out of scope
+here. Because LTX handles the reference image as a frame-0 in-place replacement
+(not a separate token group), the driving-token placement is identical for both
+:class:`DrivingMode` values in Phase 1 -- ``mode`` is stored for forward
+compatibility and to document intent, but does not yet alter the driving
+coordinates. See the plan for the deferred Phase 2 work.
+"""
+
+from __future__ import annotations
+
+from enum import Enum
+
+import torch
+
+from ltx_core.components.patchifiers import get_pixel_coords
+from ltx_core.conditioning.cond_channels import extend_cond_channels
+from ltx_core.conditioning.item import ConditioningItem
+from ltx_core.conditioning.mask_utils import update_attention_mask
+from ltx_core.tools import VideoLatentTools
+from ltx_core.types import LatentState, VideoLatentShape
+
+# Default normalization ceiling for the RoPE width axis. Must stay in sync with
+# the model's ``positional_embedding_max_pos[2]`` (see rope.py:precompute_freqs_cis
+# default ``max_pos=[20, 2048, 2048]`` and model.py `_init_` default). Driving
+# width coordinates that reach or exceed this value would wrap under RoPE.
+DEFAULT_MAX_WIDTH_POSITION = 2048
+
+
+class DrivingMode(Enum):
+    """SCAIL-2 conditioning mode. See module docstring for the Phase 1 caveat."""
+
+    ANIMATION = "animation"
+    REPLACEMENT = "replacement"
+
+
+class VideoConditionByDrivingLatent(ConditioningItem):
+    """Append driving-video tokens with a width-axis RoPE offset (SCAIL-2 ΔW).
+
+    The driving tokens are appended after the target sequence as clean latents
+    (placeholder zeros in the noisy latent), kept frozen (``denoise_mask =
+    1 - strength``), temporally aligned to the target, and shifted along the
+    width axis by ``width_offset`` so they occupy ``[ΔW, ΔW + Wv)`` while the
+    target stays at ``[0, Wv)``.
+
+    Args:
+        latent: Driving video latents ``[B, C, F, H, W]``. Must match the target
+            shape (same F/H/W) so tokens align frame-for-frame with the target.
+        mode: SCAIL-2 mode (reserved for Phase 2; see module docstring).
+        width_offset: ΔW in RoPE pixel-space width units. ``None`` (default) uses
+            the target's pixel width (``target_shape.width * scale_factors.width``),
+            placing the driving tokens immediately to the right of the target.
+        strength: 1.0 keeps the driving latent fully clean (frozen); 0.0 would
+            denoise it. Default 1.0.
+        max_width_position: RoPE width normalization ceiling; validation raises if
+            the shifted driving coordinates would reach it. Keep in sync with the
+            model's ``positional_embedding_max_pos[2]``.
+    """
+
+    def __init__(
+        self,
+        latent: torch.Tensor,
+        mode: DrivingMode = DrivingMode.ANIMATION,
+        width_offset: float | None = None,
+        strength: float = 1.0,
+        max_width_position: int = DEFAULT_MAX_WIDTH_POSITION,
+    ):
+        self.latent = latent
+        self.mode = mode
+        self.width_offset = width_offset
+        self.strength = strength
+        self.max_width_position = max_width_position
+
+    def apply_to(
+        self,
+        latent_state: LatentState,
+        latent_tools: VideoLatentTools,
+    ) -> LatentState:
+        """Append driving tokens with target-aligned time and a ΔW width shift."""
+        tokens = latent_tools.patchifier.patchify(self.latent)
+
+        num_target_tokens = latent_tools.patchifier.get_token_count(latent_tools.target_shape)
+        if tokens.shape[1] != num_target_tokens:
+            raise ValueError(
+                "VideoConditionByDrivingLatent expects the driving latent to match the target shape "
+                f"(same F/H/W): got {tokens.shape[1]} driving tokens vs {num_target_tokens} target tokens. "
+                "Resize/resample the driving video to the target resolution and frame count."
+            )
+
+        # Compute the driving tokens' own pixel-space coordinates (same flow as
+        # the base reference conditioning and create_initial_state).
+        latent_coords = latent_tools.patchifier.get_patch_grid_bounds(
+            output_shape=VideoLatentShape.from_torch_shape(self.latent.shape),
+            device=self.latent.device,
+        )
+        positions = get_pixel_coords(
+            latent_coords=latent_coords,
+            scale_factors=latent_tools.scale_factors,
+            causal_fix=latent_tools.causal_fix,
+        ).to(dtype=torch.float32)
+
+        # Temporal alignment: copy the target's time coordinates so the driving
+        # tokens sit on exactly the same time grid as z_t (robust to causal_fix /
+        # fps nuances). Token order is a flattened (f h w) grid identical to the
+        # target's, so a token-wise copy is frame-aligned.
+        positions[:, 0:1, :] = latent_state.positions[:, 0:1, :num_target_tokens].to(dtype=torch.float32)
+
+        # ΔW: shift the driving tokens along the width axis so they stay spatially
+        # detached from the target tokens. Default offset = target pixel width,
+        # giving target=[0, Wv), driving=[Wv, 2*Wv).
+        width_offset = self.width_offset
+        if width_offset is None:
+            width_offset = float(latent_tools.target_shape.width * latent_tools.scale_factors.width)
+        positions[:, 2, ...] = positions[:, 2, ...] + width_offset
+
+        max_width = positions[:, 2, ...].max().item()
+        if max_width >= self.max_width_position:
+            raise ValueError(
+                f"Driving width coordinate {max_width:.1f} reaches the RoPE ceiling "
+                f"{self.max_width_position} and would wrap. Reduce width_offset "
+                f"(currently {width_offset:.1f}) or lower the output width."
+            )
+
+        denoise_mask = torch.full(
+            size=(*tokens.shape[:2], 1),
+            fill_value=1.0 - self.strength,
+            device=self.latent.device,
+            dtype=self.latent.dtype,
+        )
+
+        new_attention_mask = update_attention_mask(
+            latent_state=latent_state,
+            attention_mask=None,
+            num_noisy_tokens=num_target_tokens,
+            num_new_tokens=tokens.shape[1],
+            batch_size=tokens.shape[0],
+            device=self.latent.device,
+            dtype=self.latent.dtype,
+        )
+
+        return LatentState(
+            latent=torch.cat([latent_state.latent, torch.zeros_like(tokens)], dim=1),
+            denoise_mask=torch.cat([latent_state.denoise_mask, denoise_mask], dim=1),
+            positions=torch.cat([latent_state.positions, positions], dim=2),
+            clean_latent=torch.cat([latent_state.clean_latent, tokens], dim=1),
+            attention_mask=new_attention_mask,
+            cond_channels=extend_cond_channels(latent_state, tokens.shape[1]),
+        )

packages/ltx-core/src/ltx_core/conditioning/types/keyframe_cond.py

@@ -1,6 +1,7 @@
 import torch
 
 from ltx_core.components.patchifiers import get_pixel_coords
+from ltx_core.conditioning.cond_channels import extend_cond_channels
 from ltx_core.conditioning.item import ConditioningItem
 from ltx_core.conditioning.mask_utils import update_attention_mask
 from ltx_core.tools import VideoLatentTools
@@ -81,4 +82,5 @@ class VideoConditionByKeyframeIndex(ConditioningItem):
             positions=torch.cat([latent_state.positions, positions], dim=2),
             clean_latent=torch.cat([latent_state.clean_latent, tokens], dim=1),
             attention_mask=new_attention_mask,
+            cond_channels=extend_cond_channels(latent_state, tokens.shape[1]),
         )

packages/ltx-core/src/ltx_core/conditioning/types/mask_channels_cond.py

@@ -0,0 +1,122 @@
+"""SCAIL-2 in-context mask conditioning (mechanism 2) for LTX-2.
+
+Encodes ``K+1`` semantic pixel-space masks (1 environment switch + ``K`` character
+binding slots) into per-token conditioning channels and writes them onto the
+sequence via :attr:`LatentState.cond_channels`. Following the paper, mask signals
+are carried by the *driving* (and reference) tokens while the noisy target keeps
+an all-zero mask -- so by default the channels are written onto the last ``N``
+tokens of the sequence (the appended driving group), which means **this item must
+be applied after the driving conditioning**.
+
+Channel expansion (LTX-2 faithful port of the paper's ``4(K+1)``): each semantic
+mask is spatially downsampled to the latent grid and temporally stacked along the
+channel dimension by the VAE temporal factor ``t`` (8 for LTX-2, vs 4 for the
+paper's Wan-2.1 backbone), giving ``t*(K+1)`` channels -- ``8*7 = 56`` for the
+default ``K=6``. The model's ``patchify_proj`` must be built with a matching
+``mask_conditioning_channels`` (see :class:`ltx_core.model.transformer.model.LTXModel`).
+"""
+
+from __future__ import annotations
+
+from dataclasses import replace
+
+import torch
+import torch.nn.functional as F
+from einops import rearrange
+
+from ltx_core.conditioning.item import ConditioningItem
+from ltx_core.tools import VideoLatentTools
+from ltx_core.types import LatentState
+
+
+def encode_mask_channels(
+    masks: torch.Tensor,
+    temporal_factor: int,
+    height_lat: int,
+    width_lat: int,
+    frames_lat: int,
+) -> torch.Tensor:
+    """Encode ``[B, K+1, F_pix, H_pix, W_pix]`` masks into ``[B, t*(K+1), F_lat, H_lat, W_lat]``.
+
+    Each semantic mask is area-downsampled to the latent spatial grid, then the pixel frames that
+    map to one latent frame are stacked along the channel dimension (temporal factor ``t``). The
+    causal first latent frame corresponds to a single pixel frame, which is replicated across its
+    ``t`` stacked channels so every latent frame yields a uniform ``t`` channels per semantic class.
+    """
+    b, s, f_pix, _h_pix, _w_pix = masks.shape
+    t = temporal_factor
+    expected_f_pix = (frames_lat - 1) * t + 1
+    if f_pix != expected_f_pix:
+        raise ValueError(
+            f"mask pixel frames ({f_pix}) incompatible with latent frames ({frames_lat}) at temporal "
+            f"factor {t}: expected (F_lat - 1) * {t} + 1 = {expected_f_pix}."
+        )
+
+    # Spatial downsample every (semantic, frame) mask to the latent grid.
+    flat = rearrange(masks.to(dtype=torch.float32), "b s f h w -> (b s f) 1 h w")
+    down = F.interpolate(flat, size=(height_lat, width_lat), mode="area")
+    down = rearrange(down, "(b s f) 1 h w -> b s f h w", b=b, s=s)
+
+    # Temporal stacking. Latent frame 0 = pixel frame 0 (causal), replicated across t channels;
+    # latent frames 1.. group t consecutive pixel frames.
+    first = down[:, :, :1].repeat(1, 1, t, 1, 1).unsqueeze(2)  # [B, S, 1, t, H, W]
+    rest = rearrange(down[:, :, 1:], "b s (fl t) h w -> b s fl t h w", t=t)  # [B, S, F_lat-1, t, H, W]
+    stacked = torch.cat([first, rest], dim=2)  # [B, S, F_lat, t, H, W]
+
+    # Fold (semantic, temporal) into a single channel axis, semantic-major: channel = s * t + ti.
+    return rearrange(stacked, "b s fl t h w -> b (s t) fl h w")  # [B, t*(K+1), F_lat, H_lat, W_lat]
+
+
+class VideoConditionByMaskChannels(ConditioningItem):
+    """Write SCAIL-2 in-context mask channels onto the driving/reference tokens.
+
+    Args:
+        masks: Pixel-space semantic masks ``[B, K+1, F_pix, H_pix, W_pix]``. Channel 0 is the
+            environment switch (whether the environment comes from the reference vs the driving
+            video); channels ``1..K`` are the character binding slots (regions sharing a slot share
+            motion). ``F_pix`` must equal ``(F_lat - 1) * temporal_factor + 1``.
+        applies_to_last_n: Number of trailing tokens to write the mask onto. ``None`` (default) uses
+            the target token count, i.e. the appended driving group when this item runs right after
+            the driving conditioning. The noisy target tokens keep an all-zero mask (paper-faithful).
+    """
+
+    def __init__(self, masks: torch.Tensor, applies_to_last_n: int | None = None):
+        self.masks = masks
+        self.applies_to_last_n = applies_to_last_n
+
+    def apply_to(self, latent_state: LatentState, latent_tools: VideoLatentTools) -> LatentState:
+        shape = latent_tools.target_shape
+        cond = encode_mask_channels(
+            masks=self.masks,
+            temporal_factor=latent_tools.scale_factors.time,
+            height_lat=shape.height,
+            width_lat=shape.width,
+            frames_lat=shape.frames,
+        )
+        tokens = latent_tools.patchifier.patchify(cond)  # [B, T_region, C_mask]
+        b, n_region, c_mask = tokens.shape
+
+        total_tokens = latent_state.latent.shape[1]
+        n = self.applies_to_last_n if self.applies_to_last_n is not None else n_region
+        if n != n_region:
+            raise ValueError(
+                f"applies_to_last_n ({n}) must equal the encoded mask token count ({n_region})."
+            )
+        if n > total_tokens:
+            raise ValueError(
+                f"cannot write {n} mask tokens onto a sequence of only {total_tokens} tokens."
+            )
+
+        cond_channels = latent_state.cond_channels
+        if cond_channels is None:
+            cond_channels = tokens.new_zeros(b, total_tokens, c_mask)
+        else:
+            if cond_channels.shape[2] != c_mask:
+                raise ValueError(
+                    f"existing cond_channels width ({cond_channels.shape[2]}) != mask channels ({c_mask})."
+                )
+            cond_channels = cond_channels.clone()
+
+        start = total_tokens - n
+        cond_channels[:, start : start + n] = tokens.to(dtype=cond_channels.dtype)
+        return replace(latent_state, cond_channels=cond_channels)

packages/ltx-core/src/ltx_core/conditioning/types/reference_audio_cond.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 
 import torch
 
+from ltx_core.conditioning.cond_channels import extend_cond_channels
 from ltx_core.conditioning.mask_utils import update_attention_mask
 from ltx_core.tools import LatentTools
 from ltx_core.types import LatentState
@@ -56,4 +57,5 @@ class AudioConditionByReferenceLatent:
             positions=torch.cat([latent_state.positions, self.positions], dim=2),
             clean_latent=torch.cat([latent_state.clean_latent, tokens], dim=1),
             attention_mask=new_attention_mask,
+            cond_channels=extend_cond_channels(latent_state, tokens.shape[1]),
         )

packages/ltx-core/src/ltx_core/conditioning/types/reference_video_cond.py

@@ -3,6 +3,7 @@
 import torch
 
 from ltx_core.components.patchifiers import get_pixel_coords
+from ltx_core.conditioning.cond_channels import extend_cond_channels
 from ltx_core.conditioning.item import ConditioningItem
 from ltx_core.conditioning.mask_utils import update_attention_mask
 from ltx_core.tools import VideoLatentTools
@@ -99,4 +100,5 @@ class VideoConditionByReferenceLatent(ConditioningItem):
             positions=torch.cat([latent_state.positions, positions], dim=2),
             clean_latent=torch.cat([latent_state.clean_latent, tokens], dim=1),
             attention_mask=new_attention_mask,
+            cond_channels=extend_cond_channels(latent_state, tokens.shape[1]),
         )

packages/ltx-core/src/ltx_core/model/transformer/mask_channels_checkpoint.py

@@ -0,0 +1,92 @@
+"""Checkpoint surgery to enable SCAIL-2 in-context mask channels on a trained LTX-2 model.
+
+Widening ``LTXModel.mask_conditioning_channels`` from 0 to ``C`` grows the video
+``patchify_proj`` weight from ``[inner, in]`` to ``[inner, in + C]``. This helper
+appends ``C`` **zero** input columns to that weight so a converted checkpoint is
+numerically identical to the original until the new columns are finetuned -- the
+extra channels contribute nothing at load time.
+
+Usage::
+
+    sd = load_state_dict(path)
+    widen_patchify_proj_for_mask_channels(sd, mask_channels=56)
+    model = LTXModel(..., mask_conditioning_channels=56)
+    model.load_state_dict(sd)
+"""
+
+from __future__ import annotations
+
+import torch
+
+_VIDEO_PROJ_SUFFIX = "patchify_proj.weight"
+_AUDIO_PROJ_SUFFIX = "audio_patchify_proj.weight"
+
+
+def widen_module_patchify_proj_for_mask_channels(model: torch.nn.Module, mask_channels: int) -> torch.nn.Module:
+    """In place, widen a live ``LTXModel``'s video ``patchify_proj`` to accept ``mask_channels`` extra inputs.
+
+    Replaces ``model.patchify_proj`` with a wider ``Linear`` whose original input columns are copied and
+    whose ``mask_channels`` new columns are zero (so the model reproduces its pre-widening output until
+    those columns are trained). Also sets ``model.mask_conditioning_channels`` for metadata/consistency.
+    Idempotent guard: raises if the model is already widened to a different value.
+    """
+    if mask_channels < 0:
+        raise ValueError(f"mask_channels must be non-negative, got {mask_channels}")
+
+    proj = getattr(model, "patchify_proj", None)
+    if not isinstance(proj, torch.nn.Linear):
+        raise AttributeError("model has no linear 'patchify_proj' to widen")
+
+    already = int(getattr(model, "mask_conditioning_channels", 0))
+    if mask_channels in (0, already):
+        model.mask_conditioning_channels = mask_channels
+        return model
+    if already != 0:
+        raise ValueError(f"patchify_proj already widened for {already} mask channels, refusing to re-widen")
+
+    new_in = proj.in_features + mask_channels
+    wider = torch.nn.Linear(new_in, proj.out_features, bias=proj.bias is not None)
+    wider = wider.to(device=proj.weight.device, dtype=proj.weight.dtype)
+    with torch.no_grad():
+        wider.weight.zero_()
+        wider.weight[:, : proj.in_features].copy_(proj.weight)
+        if proj.bias is not None:
+            wider.bias.copy_(proj.bias)
+    model.patchify_proj = wider
+    model.mask_conditioning_channels = mask_channels
+    return model
+
+
+def widen_patchify_proj_for_mask_channels(
+    state_dict: dict[str, torch.Tensor],
+    mask_channels: int,
+) -> dict[str, torch.Tensor]:
+    """In place, append ``mask_channels`` zero input columns to the video ``patchify_proj`` weight.
+
+    Only the video projection is touched (audio ``audio_patchify_proj`` is left alone). The bias,
+    if present, is unchanged. Returns the same dict for convenience. Idempotency is the caller's
+    responsibility -- calling twice widens twice.
+    """
+    if mask_channels < 0:
+        raise ValueError(f"mask_channels must be non-negative, got {mask_channels}")
+    if mask_channels == 0:
+        return state_dict
+
+    target_keys = [
+        key
+        for key in state_dict
+        if key.endswith(_VIDEO_PROJ_SUFFIX) and not key.endswith(_AUDIO_PROJ_SUFFIX)
+    ]
+    if not target_keys:
+        raise KeyError(
+            f"No '{_VIDEO_PROJ_SUFFIX}' weight found in the state dict; cannot widen for mask channels."
+        )
+
+    for key in target_keys:
+        weight = state_dict[key]  # [inner_dim, in_features]
+        if weight.ndim != 2:
+            raise ValueError(f"Expected 2-D weight for '{key}', got shape {tuple(weight.shape)}")
+        pad = weight.new_zeros(weight.shape[0], mask_channels)
+        state_dict[key] = torch.cat([weight, pad], dim=1)
+
+    return state_dict

packages/ltx-core/src/ltx_core/model/transformer/modality.py

@@ -38,6 +38,9 @@ class Modality:
             attention. ``None`` means unrestricted (full) attention between
             all tokens. Built incrementally by conditioning items; see
             :class:`~ltx_core.conditioning.types.attention_strength_wrapper.ConditioningItemAttentionStrengthWrapper`.
+        cond_channels: Optional in-context conditioning channels, shape ``(B, T, C_cond)``, aligned token-for-token
+            with ``latent``. Concatenated onto ``latent`` before the first projection (see
+            :meth:`TransformerArgsPreprocessor.prepare`); never noised. ``None`` for standard models.
     """
 
     latent: (
@@ -53,6 +56,7 @@ class Modality:
     enabled: bool = True
     context_mask: torch.Tensor | None = None
     attention_mask: torch.Tensor | None = None
+    cond_channels: torch.Tensor | None = None
 
     def split(self, sizes: list[int]) -> list[Modality]:
         """Split along the batch dimension into chunks of the given sizes."""

packages/ltx-core/src/ltx_core/model/transformer/model.py

@@ -73,6 +73,7 @@ class LTXModel(torch.nn.Module):
         caption_projection: torch.nn.Module | None = None,
         audio_caption_projection: torch.nn.Module | None = None,
         cross_attention_adaln: bool = False,
+        mask_conditioning_channels: int = 0,
     ):
         super().__init__()
         # Log the attention backends this transformer is built with. Reading the resolved
@@ -86,6 +87,10 @@ class LTXModel(torch.nn.Module):
         )
         self._enable_gradient_checkpointing = False
         self.cross_attention_adaln = cross_attention_adaln
+        # Extra per-token input channels (e.g. SCAIL-2 in-context mask channels) concatenated onto the
+        # video latent before ``patchify_proj``. 0 keeps the standard input width; >0 widens the first
+        # projection. Fed via ``Modality.cond_channels`` (see TransformerArgsPreprocessor.prepare).
+        self.mask_conditioning_channels = mask_conditioning_channels
         self.use_middle_indices_grid = use_middle_indices_grid
         self.rope_type = rope_type
         self.double_precision_rope = double_precision_rope
@@ -154,8 +159,11 @@ class LTXModel(torch.nn.Module):
         caption_projection: torch.nn.Module | None = None,
     ) -> None:
         """Initialize video-specific components."""
-        # Video input components
-        self.patchify_proj = torch.nn.Linear(in_channels, self.inner_dim, bias=True)
+        # Video input components. When ``mask_conditioning_channels > 0`` the first projection is
+        # widened to also accept the in-context conditioning channels concatenated onto the latent.
+        self.patchify_proj = torch.nn.Linear(
+            in_channels + self.mask_conditioning_channels, self.inner_dim, bias=True
+        )
         if caption_projection is not None:
             self.caption_projection = caption_projection
 

packages/ltx-core/src/ltx_core/model/transformer/model_configurator.py

@@ -69,6 +69,7 @@ class LTXModelConfigurator(ModelConfigurator[LTXModel]):
             caption_projection=caption_projection,
             audio_caption_projection=audio_caption_projection,
             cross_attention_adaln=config.get("cross_attention_adaln", False),
+            mask_conditioning_channels=config.get("mask_conditioning_channels", 0),
         )
 
 
@@ -120,6 +121,7 @@ class LTXVideoOnlyModelConfigurator(ModelConfigurator[LTXModel]):
             apply_gated_attention=config.get("apply_gated_attention", False),
             caption_projection=caption_projection,
             cross_attention_adaln=config.get("cross_attention_adaln", False),
+            mask_conditioning_channels=config.get("mask_conditioning_channels", 0),
         )
 
 

packages/ltx-core/src/ltx_core/model/transformer/transformer_args.py

@@ -201,12 +201,46 @@ class TransformerArgsPreprocessor:
         )
         return pe
 
+    def _apply_patchify_proj(self, modality: Modality) -> torch.Tensor:
+        """Project patchified latents, concatenating any in-context conditioning channels first.
+
+        When ``patchify_proj`` expects more input features than the latent provides (a model built
+        with ``mask_conditioning_channels > 0``), the extra width is filled by ``modality.cond_channels``
+        (per-token, never noised). If those channels are absent they default to zeros, so a widened
+        model still runs and, with zero-initialized new projection columns, reproduces the base model's
+        output exactly.
+        """
+        latent = modality.latent
+        expected = self.patchify_proj.in_features
+        actual = latent.shape[-1]
+        if expected != actual:
+            missing = expected - actual
+            if missing < 0:
+                raise ValueError(
+                    f"patchify_proj expects {expected} input features but the latent already has {actual}."
+                )
+            cond_channels = modality.cond_channels
+            if cond_channels is None:
+                cond_channels = latent.new_zeros(latent.shape[0], latent.shape[1], missing)
+            elif cond_channels.shape[-1] != missing:
+                raise ValueError(
+                    f"cond_channels has {cond_channels.shape[-1]} channels but patchify_proj needs {missing} "
+                    f"extra input features (latent {actual} + cond {cond_channels.shape[-1]} != {expected})."
+                )
+            elif cond_channels.shape[1] != latent.shape[1]:
+                raise ValueError(
+                    f"cond_channels token length {cond_channels.shape[1]} must match latent token length "
+                    f"{latent.shape[1]}."
+                )
+            latent = torch.cat([latent, cond_channels.to(dtype=latent.dtype)], dim=-1)
+        return self.patchify_proj(latent)
+
     def prepare(
         self,
         modality: Modality,
         cross_modality: Modality | None = None,  # noqa: ARG002
     ) -> TransformerArgs:
-        x = self.patchify_proj(modality.latent)
+        x = self._apply_patchify_proj(modality)
         batch_size = x.shape[0]
         timestep, embedded_timestep = self._prepare_timestep(
             modality.timesteps, self.adaln, batch_size, modality.latent.dtype

packages/ltx-core/src/ltx_core/tools.py

@@ -75,6 +75,9 @@ class LatentTools(Protocol):
         clean_latent = latent_state.clean_latent[:, :num_tokens]
         denoise_mask = torch.ones_like(latent_state.denoise_mask)[:, :num_tokens]
         positions = latent_state.positions[:, :, :num_tokens]
+        cond_channels = (
+            latent_state.cond_channels[:, :num_tokens] if latent_state.cond_channels is not None else None
+        )
 
         return LatentState(
             latent=latent,
@@ -82,6 +85,7 @@ class LatentTools(Protocol):
             positions=positions,
             clean_latent=clean_latent,
             attention_mask=None,
+            cond_channels=cond_channels,
         )
 
 

packages/ltx-core/src/ltx_core/types.py

@@ -193,6 +193,11 @@ class LatentState:
         clean_latent: Initial state of the latent before denoising, may include conditioning latents.
         attention_mask: Optional 2D self-attention mask of shape (B, T, T). Values in [0, 1] where 1 = full attention,
             0 = no attention. None means full attention everywhere. Built incrementally by conditioning items.
+        cond_channels: Optional in-context conditioning channels in patchified token space, shape (B, T, C_cond),
+            aligned token-for-token with ``latent``. These extra per-token feature channels (e.g. SCAIL-2 mask
+            channels) are concatenated onto the latent right before the model's first projection; they are never
+            noised or denoised. ``None`` means no conditioning channels (standard models). When present, the token
+            length T must always match ``latent``; token-appending conditioning items extend it with zeros.
     """
 
     latent: torch.Tensor
@@ -200,6 +205,7 @@ class LatentState:
     positions: torch.Tensor
     clean_latent: torch.Tensor
     attention_mask: torch.Tensor | None = None
+    cond_channels: torch.Tensor | None = None
 
     def clone(self) -> "LatentState":
         return LatentState(
@@ -208,4 +214,5 @@ class LatentState:
             positions=self.positions.clone(),
             clean_latent=self.clean_latent.clone(),
             attention_mask=self.attention_mask.clone() if self.attention_mask is not None else None,
+            cond_channels=self.cond_channels.clone() if self.cond_channels is not None else None,
         )

packages/ltx-pipelines/CLAUDE.md

@@ -25,6 +25,7 @@ Inference pipelines for LTX-2 audio-video generation. Depends on `ltx-core` for
 | `DistilledPipeline` | `distilled.py` | 2 | Distilled only | Euler | Fastest inference |
 | `ICLoraPipeline` | `ic_lora.py` | 2 | Distilled only | Euler | Video-to-video with IC-LoRA control |
 | `LipDubPipeline` | `lipdub.py` | 2 | Distilled only | Euler | Lip dubbing with IC-LoRA + audio ref conditioning |
+| `ScailAnimationPipeline` | `scail_animation.py` | 2 | Distilled + SCAIL LoRA | Euler | SCAIL-2 character animation from a driving video (driving latent concat + ΔW RoPE + in-context mask channels) |
 | `RetakePipeline` | `retake.py` | 1 | Full or distilled | Euler | Video region regeneration |
 
 ## Guidance

packages/ltx-pipelines/src/ltx_pipelines/scail_animation.py

@@ -0,0 +1,349 @@
+"""SCAIL-2 character-animation inference pipeline.
+
+Two-stage distilled video generation that animates a character from a *driving*
+video: the driving latent is concatenated into the token sequence with a RoPE
+width offset (ΔW) and in-context mask channels route motion per character. This
+mirrors the distilled pipeline (``distilled.py``) and lip-dub pipeline
+(``lipdub.py``) structure, adding the SCAIL conditioning assembly.
+
+Requirements:
+- A SCAIL-trained checkpoint whose config declares ``mask_conditioning_channels``
+  (so ``LTXModelConfigurator`` builds the widened ``patchify_proj`` automatically)
+  plus the SCAIL LoRA adapter.
+- A driving video (same target resolution) and a semantic mask tensor
+  ``[K+1, F_pix, H_pix, W_pix]`` (channel 0 = environment switch, 1..K = binding
+  slots), saved as a ``.pt`` file.
+
+The SCAIL-specific conditioning assembly is factored into
+``build_scail_conditionings`` so it can be unit-tested without the heavy models.
+"""
+
+from __future__ import annotations
+
+import logging
+from collections.abc import Iterator
+
+import torch
+
+from ltx_core.components.noisers import GaussianNoiser
+from ltx_core.conditioning import (
+    ConditioningItem,
+    DrivingMode,
+    VideoConditionByDrivingLatent,
+    VideoConditionByMaskChannels,
+)
+from ltx_core.loader import LoraPathStrengthAndSDOps
+from ltx_core.loader.registry import Registry
+from ltx_core.model.transformer.compiling import CompilationConfig
+from ltx_core.model.video_vae import TilingConfig, VideoEncoder, get_video_chunks_number
+from ltx_core.quantization import QuantizationPolicy
+from ltx_core.types import Audio, VideoPixelShape
+from ltx_pipelines.utils.allocator_trim_strategy import AllocatorTrimStrategy
+from ltx_pipelines.utils.args import ImageConditioningInput, resolve_cli_params, scail_animation_arg_parser
+from ltx_pipelines.utils.blocks import (
+    DiffusionStage,
+    ImageConditioner,
+    PromptEncoder,
+    VideoDecoder,
+    VideoUpsampler,
+)
+from ltx_pipelines.utils.constants import DISTILLED_SIGMAS, STAGE_2_DISTILLED_SIGMAS
+from ltx_pipelines.utils.denoisers import SimpleDenoiser
+from ltx_pipelines.utils.helpers import assert_resolution, combined_image_conditionings, get_device
+from ltx_pipelines.utils.media_io import decode_video_by_frame, encode_video, video_preprocess
+from ltx_pipelines.utils.types import ModalitySpec, OffloadMode
+
+
+def build_scail_conditionings(
+    driving_latent: torch.Tensor,
+    masks: torch.Tensor | None,
+    *,
+    mode: DrivingMode = DrivingMode.ANIMATION,
+    strength: float = 1.0,
+    width_offset: float | None = None,
+) -> list[ConditioningItem]:
+    """Assemble the SCAIL driving + mask-channel conditioning items (order matters).
+
+    The driving item is applied first (it appends the driving tokens at the tail); the mask item is
+    applied second, writing its channels onto those trailing driving tokens (the noisy target keeps a
+    zero mask). ``masks`` may be ``None`` to run driving-only (e.g. a model without mask channels).
+
+    Args:
+        driving_latent: Driving video latent ``[B, C, F, H, W]`` (same F/H/W as the target).
+        masks: Semantic masks ``[B, K+1, F_pix, H_pix, W_pix]`` or ``None``.
+        mode: SCAIL mode (animation/replacement).
+        strength: Driving conditioning strength (1.0 keeps it clean/frozen).
+        width_offset: ΔW in RoPE pixel-space width units (None = target pixel width).
+    """
+    conditionings: list[ConditioningItem] = [
+        VideoConditionByDrivingLatent(
+            latent=driving_latent,
+            mode=mode,
+            width_offset=width_offset,
+            strength=strength,
+        )
+    ]
+    if masks is not None:
+        conditionings.append(VideoConditionByMaskChannels(masks=masks))
+    return conditionings
+
+
+def load_masks(mask_path: str, device: torch.device, dtype: torch.dtype) -> torch.Tensor:
+    """Load a semantic mask tensor from a ``.pt`` file and shape it to ``[B, K+1, F_pix, H, W]``."""
+    masks = torch.load(mask_path, map_location=device, weights_only=True)
+    if isinstance(masks, dict):
+        masks = masks["mask"]
+    if masks.dim() == 4:  # [K+1, F, H, W] -> add batch
+        masks = masks.unsqueeze(0)
+    return masks.to(device=device, dtype=dtype)
+
+
+class ScailAnimationPipeline:
+    """Two-stage distilled SCAIL-2 character animation (driving video + in-context mask channels)."""
+
+    def __init__(
+        self,
+        distilled_checkpoint_path: str,
+        gemma_root: str,
+        spatial_upsampler_path: str,
+        scail_lora: LoraPathStrengthAndSDOps,
+        device: torch.device | None = None,
+        quantization: QuantizationPolicy | None = None,
+        registry: Registry | None = None,
+        compilation_config: CompilationConfig | None = None,
+        offload_mode: OffloadMode = OffloadMode.NONE,
+        alloc_trim_strategy: AllocatorTrimStrategy = AllocatorTrimStrategy.TRIM,
+    ) -> None:
+        self.device = device or get_device()
+        self.dtype = torch.bfloat16
+
+        self.prompt_encoder = PromptEncoder(
+            distilled_checkpoint_path,
+            gemma_root,
+            self.dtype,
+            self.device,
+            registry=registry,
+            offload_mode=offload_mode,
+            alloc_trim_strategy=alloc_trim_strategy,
+        )
+        self.image_conditioner = ImageConditioner(
+            distilled_checkpoint_path,
+            self.dtype,
+            self.device,
+            registry=registry,
+            alloc_trim_strategy=alloc_trim_strategy,
+        )
+        self.stage = DiffusionStage.from_checkpoint(
+            distilled_checkpoint_path,
+            self.dtype,
+            self.device,
+            loras=(scail_lora,),
+            quantization=quantization,
+            registry=registry,
+            compilation_config=compilation_config,
+            offload_mode=offload_mode,
+            alloc_trim_strategy=alloc_trim_strategy,
+        )
+        self.upsampler = VideoUpsampler(
+            distilled_checkpoint_path,
+            spatial_upsampler_path,
+            self.dtype,
+            self.device,
+            registry=registry,
+            alloc_trim_strategy=alloc_trim_strategy,
+        )
+        self.video_decoder = VideoDecoder(
+            distilled_checkpoint_path,
+            self.dtype,
+            self.device,
+            registry=registry,
+            alloc_trim_strategy=alloc_trim_strategy,
+        )
+
+    def _encode_driving_latent(
+        self,
+        driving_video_path: str,
+        height: int,
+        width: int,
+        num_frames: int,
+        video_encoder: VideoEncoder,
+        tiling_config: TilingConfig | None,
+    ) -> torch.Tensor:
+        """Decode + encode the driving video into a VAE latent at the target resolution."""
+        frame_gen = decode_video_by_frame(path=driving_video_path, frame_cap=num_frames, device=self.device)
+        video = video_preprocess(frame_gen, height, width, self.dtype, self.device)
+        if tiling_config is not None:
+            return video_encoder.tiled_encode(video, tiling_config)
+        return video_encoder(video)
+
+    def _video_conditionings(
+        self,
+        images: list[ImageConditioningInput],
+        driving_video_path: str,
+        masks: torch.Tensor | None,
+        height: int,
+        width: int,
+        num_frames: int,
+        mode: DrivingMode,
+        driving_strength: float,
+        video_encoder: VideoEncoder,
+        tiling_config: TilingConfig | None,
+    ) -> list[ConditioningItem]:
+        conditionings = combined_image_conditionings(
+            images=images,
+            height=height,
+            width=width,
+            video_encoder=video_encoder,
+            dtype=self.dtype,
+            device=self.device,
+        )
+        driving_latent = self._encode_driving_latent(
+            driving_video_path, height, width, num_frames, video_encoder, tiling_config
+        )
+        conditionings.extend(
+            build_scail_conditionings(driving_latent, masks, mode=mode, strength=driving_strength)
+        )
+        return conditionings
+
+    @torch.inference_mode()
+    def __call__(  # noqa: PLR0913
+        self,
+        prompt: str,
+        seed: int,
+        height: int,
+        width: int,
+        num_frames: int,
+        frame_rate: float,
+        driving_video_path: str,
+        mask_path: str | None,
+        images: list[ImageConditioningInput] | None = None,
+        mode: DrivingMode = DrivingMode.ANIMATION,
+        driving_strength: float = 1.0,
+        enhance_prompt: bool = False,
+        tiling_config: TilingConfig | None = None,
+        stage_1_sigmas: torch.Tensor = DISTILLED_SIGMAS,
+        stage_2_sigmas: torch.Tensor = STAGE_2_DISTILLED_SIGMAS,
+    ) -> tuple[Iterator[torch.Tensor], Audio | None]:
+        assert_resolution(height=height, width=width, is_two_stage=True)
+        images = images or []
+
+        generator = torch.Generator(device=self.device).manual_seed(seed)
+        noiser = GaussianNoiser(generator=generator)
+        encode_tiling = TilingConfig.default()
+
+        (ctx_p,) = self.prompt_encoder(
+            [prompt],
+            enhance_first_prompt=enhance_prompt,
+            enhance_prompt_image=images[0][0] if len(images) > 0 else None,
+            enhance_prompt_seed=seed,
+        )
+        video_context, audio_context = ctx_p.video_encoding, ctx_p.audio_encoding
+
+        masks = load_masks(mask_path, self.device, self.dtype) if mask_path is not None else None
+
+        def build_video_conditionings(out_height: int, out_width: int) -> list[ConditioningItem]:
+            return self.image_conditioner(
+                lambda enc: self._video_conditionings(
+                    images=images,
+                    driving_video_path=driving_video_path,
+                    masks=masks,
+                    height=out_height,
+                    width=out_width,
+                    num_frames=num_frames,
+                    mode=mode,
+                    driving_strength=driving_strength,
+                    video_encoder=enc,
+                    tiling_config=encode_tiling,
+                )
+            )
+
+        # Stage 1: low resolution.
+        stage_1_sigmas_t = stage_1_sigmas.to(dtype=torch.float32, device=self.device)
+        stage_1_conditionings = build_video_conditionings(height // 2, width // 2)
+        video_state, _ = self.stage(
+            denoiser=SimpleDenoiser(video_context, audio_context),
+            sigmas=stage_1_sigmas_t,
+            noiser=noiser,
+            width=width // 2,
+            height=height // 2,
+            frames=num_frames,
+            fps=frame_rate,
+            video=ModalitySpec(context=video_context, conditionings=stage_1_conditionings),
+            audio=ModalitySpec(context=audio_context),
+        )
+
+        # Stage 2: upsample + refine.
+        upscaled = self.upsampler(video_state.latent[:1])
+        stage_2_sigmas_t = stage_2_sigmas.to(dtype=torch.float32, device=self.device)
+        stage_2_conditionings = build_video_conditionings(height, width)
+        video_state, _ = self.stage(
+            denoiser=SimpleDenoiser(video_context, audio_context),
+            sigmas=stage_2_sigmas_t,
+            noiser=noiser,
+            width=width,
+            height=height,
+            frames=num_frames,
+            fps=frame_rate,
+            video=ModalitySpec(
+                context=video_context,
+                conditionings=stage_2_conditionings,
+                noise_scale=stage_2_sigmas_t[0].item(),
+                initial_latent=upscaled,
+            ),
+            audio=ModalitySpec(context=audio_context),
+        )
+
+        decoded_video = self.video_decoder(video_state.latent, tiling_config, generator)
+        return decoded_video, None
+
+
+@torch.inference_mode()
+def main() -> None:
+    logging.basicConfig(level=logging.INFO)
+    params = resolve_cli_params(distilled=True)
+    parser = scail_animation_arg_parser(params=params)
+    args = parser.parse_args()
+
+    if not args.lora or len(args.lora) != 1:
+        raise ValueError("SCAIL animation requires exactly one --lora (the SCAIL adapter).")
+
+    pipeline = ScailAnimationPipeline(
+        distilled_checkpoint_path=args.distilled_checkpoint_path,
+        gemma_root=args.gemma_root,
+        spatial_upsampler_path=args.spatial_upsampler_path,
+        scail_lora=args.lora[0],
+        quantization=args.quantization,
+        compilation_config=args.compile,
+        offload_mode=args.offload_mode,
+    )
+    tiling_config = TilingConfig.default()
+    output_shape = VideoPixelShape(
+        batch=1, frames=args.num_frames, width=args.width, height=args.height, fps=args.frame_rate
+    )
+    video_chunks_number = get_video_chunks_number(output_shape.frames, tiling_config)
+    video, _audio = pipeline(
+        prompt=args.prompt,
+        seed=args.seed,
+        height=args.height,
+        width=args.width,
+        num_frames=args.num_frames,
+        frame_rate=args.frame_rate,
+        driving_video_path=args.driving_video,
+        mask_path=args.mask_path,
+        images=args.images if hasattr(args, "images") else [],
+        mode=DrivingMode(args.mode),
+        driving_strength=args.driving_strength,
+        enhance_prompt=args.enhance_prompt,
+        tiling_config=tiling_config,
+    )
+    encode_video(
+        video=video,
+        fps=int(args.frame_rate),
+        audio=None,
+        output_path=args.output_path,
+        video_chunks_number=video_chunks_number,
+    )
+
+
+if __name__ == "__main__":
+    main()

packages/ltx-pipelines/src/ltx_pipelines/utils/args.py

@@ -809,3 +809,40 @@ def default_2_stage_distilled_arg_parser(params: PipelineParams = LTX_2_3_PARAMS
         ),
     )
     return parser
+
+
+def scail_animation_arg_parser(params: PipelineParams = LTX_2_3_PARAMS) -> argparse.ArgumentParser:
+    """Argument parser for the SCAIL-2 character-animation pipeline (distilled, two-stage).
+
+    Extends the standard two-stage distilled parser (checkpoint paths, --prompt, --num-frames,
+    --frame-rate, --height, --width, --image, --lora, --spatial-upsampler-path) with the SCAIL
+    driving/mask conditioning inputs. The single --lora is the SCAIL adapter.
+    """
+    parser = default_2_stage_distilled_arg_parser(params=params)
+    parser.add_argument(
+        "--driving-video",
+        type=resolve_existing_path,
+        required=True,
+        help="Driving video file whose motion is transferred to the animated character.",
+    )
+    parser.add_argument(
+        "--mask-path",
+        type=resolve_existing_path,
+        default=None,
+        help="Optional .pt file with semantic masks [K+1, F_pix, H, W] (ch0 = environment switch, "
+        "1..K = character binding slots). Omit for a driving-only (mask-free) model.",
+    )
+    parser.add_argument(
+        "--mode",
+        type=str,
+        choices=["animation", "replacement"],
+        default="animation",
+        help="SCAIL conditioning mode (default: animation).",
+    )
+    parser.add_argument(
+        "--driving-strength",
+        type=float,
+        default=1.0,
+        help="Driving conditioning strength; 1.0 keeps the driving latent clean/frozen (default: 1.0).",
+    )
+    return parser

packages/ltx-pipelines/src/ltx_pipelines/utils/helpers.py

@@ -269,6 +269,7 @@ def modality_from_latent_state(
         context=context,
         context_mask=None,
         attention_mask=state.attention_mask,
+        cond_channels=state.cond_channels,
     )
 
 

packages/ltx-trainer/configs/README.md

@@ -23,5 +23,10 @@ adjust paths, dataset, and hyperparameters.
 | **Audio Inpainting**  | —         | Generated | `mask`              | [`audio_inpainting_lora.yaml`](./audio_inpainting_lora.yaml) |
 | **A2A IC-LoRA**       | —         | Generated | `reference`         | [`a2a_ic_lora.yaml`](./a2a_ic_lora.yaml) |
 | **AV2AV IC-LoRA**     | Generated | Generated | `reference` (both)  | [`av2av_ic_lora.yaml`](./av2av_ic_lora.yaml) |
+| **SCAIL Animation**   | Generated | —         | `driving` + `mask_channels` | [`scail_animation_lora.yaml`](./scail_animation_lora.yaml) |
 
 The [`accelerate/`](./accelerate) directory holds the Accelerate launch configs (FSDP, DDP) for multi-GPU training.
+
+> **SCAIL Animation** (SCAIL-2 character animation) also sets `model.mask_conditioning_channels: 56` to widen the
+> video `patchify_proj` for the in-context mask channels; in LoRA mode that projection is unfrozen so the new columns
+> train. See [Training Modes Guide](../docs/training-modes.md).

packages/ltx-trainer/configs/scail_animation_lora.yaml

@@ -0,0 +1,132 @@
+# =============================================================================
+# LTX-2 SCAIL-2 Character Animation (LoRA + mask channels) Training Configuration
+# =============================================================================
+#
+# Trains SCAIL-2-style end-to-end character animation: a driving video latent is
+# concatenated into the token sequence with a RoPE width offset (ΔW), and
+# in-context mask channels (1 environment switch + K binding slots) are attached
+# to the driving tokens to route motion per character.
+#
+# This combines LoRA on the attention/FFN blocks with an *unfrozen* widened
+# patchify_proj (its new mask-channel input columns cannot be reached by LoRA and
+# are trained directly). Set `model.mask_conditioning_channels` to the channel
+# count = temporal_factor * (K + 1) = 8 * (6 + 1) = 56 for the default K=6.
+#
+# Dataset structure:
+#   preprocessed_data_root/
+#   ├── latents/           # Target video latents (what the model generates)
+#   ├── conditions/        # Text embeddings for each video
+#   ├── driving_latents/   # Driving video latents (same F/H/W as target)
+#   └── char_masks/        # Semantic masks per sample, "mask" = [K+1, F_pix, H_pix, W_pix]
+#                          #   channel 0 = environment switch, 1..K = character binding slots
+#
+# =============================================================================
+
+model:
+  model_path: "path/to/ltx-2-model.safetensors"
+  text_encoder_path: "path/to/gemma-text-encoder"
+  training_mode: "lora"
+  # Widen the video patchify_proj by 56 zero-init input columns (8 * (K+1), K=6).
+  # In LoRA mode the trainer additionally unfreezes patchify_proj so these train.
+  mask_conditioning_channels: 56
+  load_checkpoint: null
+
+lora:
+  rank: 32
+  alpha: 32
+  dropout: 0.0
+  target_modules:
+    - "attn1.to_k"
+    - "attn1.to_q"
+    - "attn1.to_v"
+    - "attn1.to_out.0"
+    - "attn2.to_k"
+    - "attn2.to_q"
+    - "attn2.to_v"
+    - "attn2.to_out.0"
+    - "ff.net.0.proj"
+    - "ff.net.2"
+
+training_strategy:
+  name: "flexible"
+  video:
+    is_generated: true
+    latents_dir: "latents"
+    conditions:
+      # SCAIL-2 driving conditioning: concatenate driving latents with a RoPE width
+      # offset so they stay spatially detached from the target tokens.
+      - type: driving
+        latents_dir: "driving_latents"
+        mode: "animation"        # "animation" | "replacement"
+        width_offset: null        # null = target pixel width (driving sits just to the right)
+        probability: 1.0
+      # In-context mask channels attached to the driving tokens (target stays zero-mask).
+      - type: mask_channels
+        mask_dir: "char_masks"
+        num_slots: 6              # K binding slots; channels = 8 * (K + 1) = 56 (match model.mask_conditioning_channels)
+
+optimization:
+  learning_rate: 2e-4
+  steps: 3000
+  batch_size: 1
+  gradient_accumulation_steps: 1
+  max_grad_norm: 1.0
+  optimizer_type: "adamw"
+  scheduler_type: "linear"
+  scheduler_params: { }
+  enable_gradient_checkpointing: true
+
+acceleration:
+  mixed_precision_mode: "bf16"
+  quantization: null
+  load_text_encoder_in_8bit: false
+  offload_optimizer_during_validation: false
+
+data:
+  preprocessed_data_root: "/path/to/preprocessed/data"
+  num_dataloader_workers: 2
+
+validation:
+  # NOTE: SCAIL driving + mask-channel validation conditions are not yet wired into
+  # the validation runner (Phase 3 training path only). Keep validation minimal /
+  # disabled until the inference pipeline (Phase 4) lands.
+  samples:
+    - prompt: >-
+        A person performing an energetic dance routine, matching the motion of the driving
+        performer, with crisp footwork and expressive arm movements in a bright studio.
+      conditions: []
+  negative_prompt: "worst quality, inconsistent motion, blurry, jittery, distorted"
+  video_dims: [ 512, 512, 81 ]
+  frame_rate: 25.0
+  seed: 42
+  inference_steps: 30
+  interval: null   # disabled: driving/mask validation lands in Phase 4
+  guidance_scale: 4.0
+  stg_scale: 1.0
+  stg_blocks: [29]
+  stg_mode: "stg_v"
+  generate_audio: false
+  skip_initial_validation: true
+
+checkpoints:
+  interval: 250
+  keep_last_n: 3
+  precision: "bfloat16"
+
+flow_matching:
+  timestep_sampling_mode: "shifted_logit_normal"
+  timestep_sampling_params: { }
+
+hub:
+  push_to_hub: false
+  hub_model_id: null
+
+wandb:
+  enabled: false
+  project: "ltx-2-trainer"
+  entity: null
+  tags: [ "ltx2", "scail-2", "character-animation" ]
+  log_validation_videos: true
+
+seed: 42
+output_dir: "outputs/scail_animation_lora"

packages/ltx-trainer/docs/training-modes.md

@@ -40,6 +40,7 @@ Before diving into individual modes, here are the core ideas behind the flexible
 | **Audio Inpainting**  | —         | Generated | `mask`              | [`audio_inpainting_lora`](../configs/audio_inpainting_lora.yaml) |
 | **A2A IC-LoRA**       | —         | Generated | `reference`         | [`a2a_ic_lora`](../configs/a2a_ic_lora.yaml) |
 | **AV2AV IC-LoRA**     | Generated | Generated | `reference` (both)  | [`av2av_ic_lora`](../configs/av2av_ic_lora.yaml) |
+| **SCAIL Animation**   | Generated | —         | `driving` + `mask_channels` | [`scail_animation_lora`](../configs/scail_animation_lora.yaml) |
 
 ---
 

packages/ltx-trainer/pyproject.toml

@@ -54,7 +54,7 @@ build-backend = "hatchling.build"
 
 
 [tool.ruff]
-target-version = "1.1.7"
+target-version = "py310"
 line-length = 120
 # Restrict isort first-party detection to src/ so stray dirs (e.g. wandb/ run output)
 # next to pyproject.toml don't get classified as first-party packages. See ruff#10519.

packages/ltx-trainer/scripts/process_char_masks.py

@@ -0,0 +1,217 @@
+#!/usr/bin/env python3
+
+"""Preprocess SCAIL-2 character binding-slot masks into per-sample ``.pt`` tensors.
+
+Each sample's input is a **label-map** video or image whose integer pixel values index
+characters: ``0`` = environment/background, ``k`` in ``1..K`` = character *k* (assigned to
+binding slot *k*). This script aligns the label map to the target video's latent grid (read
+from the saved video-latent metadata), and emits the pixel-space semantic-mask tensor that
+``ltx_core.conditioning.encode_mask_channels`` (and the trainer's ``mask_channels`` condition)
+consumes:
+
+    {"mask": tensor[K+1, F_pix, H_pix, W_pix]}   # ch0 = environment switch, ch1..K = binding slots
+
+where ``F_pix = (latent_frames - 1) * 8 + 1``, ``H_pix = latent_h * 32``, ``W_pix = latent_w * 32``
+(the SCAIL encoder then spatially downsamples + temporally stacks these to the 8*(K+1)=56 channels).
+
+Label maps are resized with **nearest-neighbour** interpolation so integer labels are never
+blended. The environment-switch channel (ch0) is filled uniformly with ``--environment-switch``
+(``0.0`` = derive the environment from the reference image, ``1.0`` = from the driving video),
+matching the paper's single-bit environment signal.
+
+Output naming mirrors the target latents (same relative path), so ``char_masks/`` lines up
+file-for-file with ``latents/`` / ``driving_latents/`` for the trainer's ``PrecomputedDataset``.
+
+Standalone usage::
+
+    python scripts/process_char_masks.py dataset.csv \\
+        --mask-column char_labels --latents-dir ./latents --output-dir ./char_masks --num-slots 6
+"""
+
+from pathlib import Path
+
+import numpy as np
+import torch
+import typer
+from PIL import Image
+
+# Sibling scripts (resolved via scripts/ on sys.path), reused so naming + alignment match the other stages.
+from process_videos import (
+    IMAGE_FILE_EXTENSIONS,
+    VAE_SPATIAL_FACTOR,
+    VAE_TEMPORAL_FACTOR,
+    _atomic_save,
+    _load_paths_from_dataset,
+    _output_relative,
+)
+from rich.console import Console
+from rich.progress import BarColumn, MofNCompleteColumn, Progress, SpinnerColumn, TextColumn, TimeElapsedColumn
+
+from ltx_trainer import logger
+from ltx_trainer.video_utils import read_video
+
+app = typer.Typer(
+    pretty_exceptions_enable=False,
+    no_args_is_help=True,
+    help="Preprocess SCAIL-2 character label maps into [K+1, F_pix, H_pix, W_pix] mask tensors.",
+)
+
+
+def _load_label_frames(mask_file: Path, pixel_f: int) -> torch.Tensor:
+    """Load a label-map video/image as integer labels ``[F_pix, H, W]`` (float-typed integers).
+
+    Images are tiled across ``pixel_f`` frames. Videos are read via the shared ``read_video`` helper
+    (which returns ``[F, C, H, W]`` in ``[0, 1]``); labels are recovered as ``round(value * 255)``,
+    so label maps must be stored as small-integer grayscale (label ``k`` -> pixel value ``k``).
+    """
+    if mask_file.suffix.lower() in IMAGE_FILE_EXTENSIONS:
+        arr = np.array(Image.open(mask_file).convert("L"))  # exact integer labels [H, W] (writable copy)
+        labels = torch.from_numpy(arr).float()
+        return labels.unsqueeze(0).expand(pixel_f, -1, -1).contiguous()
+
+    frames, _ = read_video(str(mask_file), max_frames=pixel_f)  # [F, C, H, W] in [0, 1]
+    labels = frames[:, 0].mul(255.0).round()  # channel 0 -> integer labels [F, H, W]
+    if labels.shape[0] < pixel_f:
+        # Pad by repeating the last frame so every latent frame has a label map.
+        pad = labels[-1:].expand(pixel_f - labels.shape[0], -1, -1)
+        labels = torch.cat([labels, pad], dim=0)
+    return labels[:pixel_f]
+
+
+def _labels_to_slot_masks(labels: torch.Tensor, num_slots: int, environment_switch: float) -> torch.Tensor:
+    """Convert integer label frames ``[F, H, W]`` to ``[K+1, F, H, W]`` (ch0 env switch, ch1..K slots)."""
+    f_pix, h, w = labels.shape
+    out = torch.zeros(num_slots + 1, f_pix, h, w, dtype=torch.float32)
+    out[0] = environment_switch  # uniform environment-switch channel
+    for k in range(1, num_slots + 1):
+        out[k] = (labels == k).float()
+    if bool(((labels > num_slots) & (labels > 0)).any().item()):
+        logger.warning(
+            f"Label map contains ids > num_slots ({num_slots}); those pixels are dropped (treated as environment)."
+        )
+    return out
+
+
+def _resize_labels(labels: torch.Tensor, pixel_h: int, pixel_w: int) -> torch.Tensor:
+    """Nearest-neighbour resize of integer label frames ``[F, H, W]`` to ``[F, pixel_h, pixel_w]``."""
+    if labels.shape[1:] == (pixel_h, pixel_w):
+        return labels
+    return torch.nn.functional.interpolate(
+        labels.unsqueeze(1), size=(pixel_h, pixel_w), mode="nearest"
+    ).squeeze(1)
+
+
+def compute_char_masks(
+    dataset_file: str | Path,
+    mask_column: str,
+    latents_dir: str,
+    output_dir: str,
+    num_slots: int = 6,
+    environment_switch: float = 0.0,
+    main_media_column: str | None = None,
+    overwrite: bool = False,
+) -> None:
+    """Preprocess character label maps into ``[K+1, F_pix, H_pix, W_pix]`` mask tensors.
+
+    Args:
+        dataset_file: Metadata file (CSV/JSON/JSONL) with a column of label-map paths.
+        mask_column: Column containing the per-sample label-map video/image paths.
+        latents_dir: Directory of target video latents (read for spatial/temporal alignment).
+        output_dir: Directory to write ``char_masks`` ``.pt`` files.
+        num_slots: Number of binding slots K (output has ``K+1`` channels).
+        environment_switch: Uniform value for ch0 (0.0 = env from reference, 1.0 = from driving video).
+        main_media_column: Column used for output naming (defaults to ``mask_column``); set it to the
+            target-video column so masks align with ``latents/`` when label maps live elsewhere.
+        overwrite: Recompute even if the output already exists.
+    """
+    dataset_path = Path(dataset_file)
+    data_root = dataset_path.parent
+    latents_path = Path(latents_dir)
+    output_path = Path(output_dir)
+    output_path.mkdir(parents=True, exist_ok=True)
+
+    naming_column = main_media_column or mask_column
+    mask_paths = _load_paths_from_dataset(dataset_path, mask_column)
+    naming_paths = _load_paths_from_dataset(dataset_path, naming_column) if naming_column != mask_column else mask_paths
+
+    console = Console()
+    success = 0
+    with Progress(
+        SpinnerColumn(),
+        TextColumn("[progress.description]{task.description}"),
+        BarColumn(),
+        MofNCompleteColumn(),
+        TimeElapsedColumn(),
+        console=console,
+    ) as progress:
+        task = progress.add_task("Processing char masks", total=len(mask_paths))
+        for mask_file, naming_file in zip(mask_paths, naming_paths, strict=True):
+            progress.advance(task)
+            rel_path = _output_relative(naming_file, data_root)
+            latent_file = latents_path / rel_path.with_suffix(".pt")
+            out_file = output_path / rel_path.with_suffix(".pt")
+
+            if not latent_file.exists():
+                logger.warning(f"No target latent at {latent_file}, skipping mask {mask_file}")
+                continue
+            if not overwrite and out_file.is_file():
+                continue
+
+            meta = torch.load(latent_file, map_location="cpu", weights_only=True)
+            pixel_h = meta["height"] * VAE_SPATIAL_FACTOR
+            pixel_w = meta["width"] * VAE_SPATIAL_FACTOR
+            pixel_f = (meta["num_frames"] - 1) * VAE_TEMPORAL_FACTOR + 1
+
+            labels = _load_label_frames(mask_file, pixel_f)
+            labels = _resize_labels(labels, pixel_h, pixel_w)
+            mask = _labels_to_slot_masks(labels, num_slots, environment_switch)
+
+            out_file.parent.mkdir(parents=True, exist_ok=True)
+            _atomic_save({"mask": mask.contiguous()}, out_file)
+            success += 1
+
+    logger.info(f"Char-mask preprocessing complete: {success} masks saved to {output_path}")
+
+
+@app.command()
+def main(
+    dataset_file: str = typer.Argument(..., help="Metadata file (CSV/JSON/JSONL) with a label-map column"),
+    mask_column: str = typer.Option(..., help="Column of per-sample label-map video/image paths"),
+    latents_dir: str = typer.Option(..., help="Directory of target video latents (for alignment)"),
+    output_dir: str = typer.Option(..., help="Output directory for char_masks .pt files"),
+    num_slots: int = typer.Option(6, help="Number of character binding slots K (channels = K+1)"),
+    environment_switch: float = typer.Option(
+        0.0, help="Uniform ch0 value: 0.0 = environment from reference, 1.0 = from driving video"
+    ),
+    main_media_column: str | None = typer.Option(
+        None, help="Column for output naming (defaults to --mask-column; set to the target-video column to align)"
+    ),
+    overwrite: bool = typer.Option(False, help="Recompute even if the output already exists"),
+) -> None:
+    """Preprocess SCAIL-2 character label maps into ``[K+1, F_pix, H_pix, W_pix]`` mask tensors.
+
+    Example::
+
+        python scripts/process_char_masks.py dataset.csv \\
+            --mask-column char_labels --latents-dir ./latents \\
+            --output-dir ./char_masks --num-slots 6 --main-media-column media_path
+    """
+    if not Path(dataset_file).is_file():
+        raise typer.BadParameter(f"Dataset file not found: {dataset_file}")
+    if num_slots < 1:
+        raise typer.BadParameter("--num-slots must be >= 1")
+
+    compute_char_masks(
+        dataset_file=dataset_file,
+        mask_column=mask_column,
+        latents_dir=latents_dir,
+        output_dir=output_dir,
+        num_slots=num_slots,
+        environment_switch=environment_switch,
+        main_media_column=main_media_column,
+        overwrite=overwrite,
+    )
+
+
+if __name__ == "__main__":
+    app()

packages/ltx-trainer/src/ltx_trainer/config.py

@@ -252,6 +252,14 @@ class ModelConfig(ConfigBaseModel):
         description="Training mode - either LoRA fine-tuning or full model fine-tuning",
     )
 
+    mask_conditioning_channels: int = Field(
+        default=0,
+        ge=0,
+        description="SCAIL-2 in-context mask channels. If > 0, the video patchify_proj is widened by this "
+        "many zero-init input columns at load time. Use with a 'driving' + 'mask_channels' condition and, "
+        "in LoRA mode, patchify_proj is additionally unfrozen so the new columns can train. 0 disables.",
+    )
+
     load_checkpoint: str | Path | None = Field(
         default=None,
         description="Path to a checkpoint file or directory to load from. "

packages/ltx-trainer/src/ltx_trainer/model_loader.py

@@ -50,27 +50,39 @@ def load_transformer(
     checkpoint_path: str | Path,
     device: Device = "cpu",
     dtype: torch.dtype = torch.bfloat16,
+    mask_conditioning_channels: int = 0,
 ) -> "LTXModel":
     """Load the LTX transformer model.
     Args:
         checkpoint_path: Path to the safetensors checkpoint file
         device: Device to load model on
         dtype: Data type for model weights
+        mask_conditioning_channels: If > 0, widen the video ``patchify_proj`` by this many zero-init
+            input columns after loading (SCAIL-2 in-context mask channels). The converted model
+            reproduces the base output exactly until those columns are trained.
     Returns:
         Loaded LTXModel transformer
     """
     from ltx_core.loader.single_gpu_model_builder import SingleGPUModelBuilder
+    from ltx_core.model.transformer.mask_channels_checkpoint import (
+        widen_module_patchify_proj_for_mask_channels,
+    )
     from ltx_core.model.transformer.model_configurator import (
         LTXV_MODEL_COMFY_RENAMING_MAP,
         LTXModelConfigurator,
     )
 
-    return SingleGPUModelBuilder(
+    model = SingleGPUModelBuilder(
         model_path=str(checkpoint_path),
         model_class_configurator=LTXModelConfigurator,
         model_sd_ops=LTXV_MODEL_COMFY_RENAMING_MAP,
     ).build(device=_to_torch_device(device), dtype=dtype)
 
+    if mask_conditioning_channels > 0:
+        widen_module_patchify_proj_for_mask_channels(model, mask_conditioning_channels)
+
+    return model
+
 
 def load_video_vae_encoder(
     checkpoint_path: str | Path,

packages/ltx-trainer/src/ltx_trainer/trainer.py

@@ -399,6 +399,7 @@ class LtxvTrainer:
             checkpoint_path=self._config.model.model_path,
             device="cpu",
             dtype=torch.bfloat16,
+            mask_conditioning_channels=self._config.model.mask_conditioning_channels,
         )
 
         # DDP-safe: LOCAL_RANK is set by accelerate before trainer init. Loading on bare
@@ -440,9 +441,25 @@ class LtxvTrainer:
         else:
             raise ValueError(f"Unknown training mode: {self._config.model.training_mode}")
 
+        # SCAIL-2: the widened patchify_proj has new mask-channel input columns that LoRA cannot reach
+        # (they are new base parameters, not a low-rank delta on an existing weight). Unfreeze the whole
+        # patchify_proj so those columns train alongside the LoRA adapters. Harmless in full mode (already
+        # trainable). Placed before trainable-param collection so the params below pick it up.
+        if self._config.model.mask_conditioning_channels > 0:
+            self._unfreeze_patchify_proj()
+
         self._trainable_params = [p for p in self._transformer.parameters() if p.requires_grad]
         logger.debug(f"Trainable params count: {sum(p.numel() for p in self._trainable_params):,}")
 
+    def _unfreeze_patchify_proj(self) -> None:
+        """Make the video ``patchify_proj`` parameters trainable (for SCAIL-2 mask-channel columns)."""
+        count = 0
+        for name, param in self._transformer.named_parameters():
+            if "patchify_proj" in name and "audio_patchify_proj" not in name:
+                param.requires_grad_(True)
+                count += param.numel()
+        logger.info(f"Unfroze video patchify_proj for mask-channel training ({count:,} params)")
+
     def _init_timestep_sampler(self) -> None:
         """Initialize the timestep sampler based on the config."""
         sampler_cls = SAMPLERS[self._config.flow_matching.timestep_sampling_mode]

packages/ltx-trainer/src/ltx_trainer/training_strategies/flexible.py

@@ -15,6 +15,7 @@ import torch
 from pydantic import BaseModel, ConfigDict, Field, model_validator
 from torch import Tensor
 
+from ltx_core.conditioning import encode_mask_channels
 from ltx_core.model.transformer.modality import Modality
 from ltx_trainer.timestep_samplers import TimestepSampler
 from ltx_trainer.training_strategies.base_strategy import (
@@ -107,6 +108,45 @@ class ReferenceConditionConfig(BaseModel):
     probability: float = Field(default=1.0, ge=0.0, le=1.0, description="Probability of applying this condition")
 
 
+class DrivingConditionConfig(BaseModel):
+    """SCAIL-2 driving-video conditioning (concatenation with a RoPE width offset).
+    Driving latents are concatenated to the sequence like a reference, but their RoPE width
+    coordinates are shifted by ``width_offset`` (the paper's ΔW) so they stay spatially detached
+    from the target tokens. Driving tokens are clean (timestep=0) and excluded from loss.
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    type: Literal["driving"] = "driving"
+    latents_dir: str = Field(..., description="Directory for driving-video latents (same F/H/W as target)")
+    mode: Literal["animation", "replacement"] = Field(
+        default="animation",
+        description="SCAIL mode. Reserved for the reference/height-shift distinction; driving-token "
+        "placement is currently identical for both (see ltx-core VideoConditionByDrivingLatent).",
+    )
+    width_offset: float | None = Field(
+        default=None,
+        description="ΔW in RoPE pixel-space width units. None = target pixel width (driving sits just to the right).",
+    )
+    probability: float = Field(default=1.0, ge=0.0, le=1.0, description="Probability of applying this condition")
+
+
+class MaskChannelsConditionConfig(BaseModel):
+    """SCAIL-2 in-context mask channels attached to the driving tokens.
+    Encodes ``num_slots + 1`` semantic pixel masks (1 environment switch + K binding slots) into
+    ``temporal_factor * (num_slots + 1)`` per-token conditioning channels (56 for K=6 on LTX-2) and
+    writes them onto the driving tokens via ``Modality.cond_channels``. The noisy target keeps an
+    all-zero mask (paper-faithful). Requires a ``driving`` condition and a model built with a matching
+    ``mask_conditioning_channels``.
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    type: Literal["mask_channels"] = "mask_channels"
+    mask_dir: str = Field(..., description="Directory of semantic masks [K+1, F_pix, H_pix, W_pix] per sample")
+    num_slots: int = Field(default=6, ge=1, description="Number of character binding slots K (channels = t*(K+1))")
+
+
 # Discriminated union for condition configs
 ConditionConfig = Annotated[
     Union[
@@ -116,6 +156,8 @@ ConditionConfig = Annotated[
         SpatialCropConditionConfig,
         MaskConditionConfig,
         ReferenceConditionConfig,
+        DrivingConditionConfig,
+        MaskChannelsConditionConfig,
     ],
     Field(discriminator="type"),
 ]
@@ -200,9 +242,9 @@ class FlexibleStrategyConfig(TrainingStrategyConfigBase):
             if modality_config is None:
                 continue
             for cond in modality_config.conditions:
-                if isinstance(cond, ReferenceConditionConfig):
+                if isinstance(cond, (ReferenceConditionConfig, DrivingConditionConfig)):
                     sources[cond.latents_dir] = cond.latents_dir
-                elif isinstance(cond, MaskConditionConfig):
+                elif isinstance(cond, (MaskConditionConfig, MaskChannelsConditionConfig)):
                     sources[cond.mask_dir] = cond.mask_dir
 
         return sources
@@ -428,6 +470,19 @@ class FlexibleStrategy(TrainingStrategy):
                     modality_key=modality_key,
                 )
 
+        # Step 5b: Apply SCAIL-2 driving + in-context mask conditioning (video only).
+        noisy_latents, positions, timesteps, loss_mask, cond_channels = self._apply_scail_conditions(
+            modality_config=modality_config,
+            modality_key=modality_key,
+            noisy_latents=noisy_latents,
+            positions=positions,
+            timesteps=timesteps,
+            loss_mask=loss_mask,
+            data=data,
+            batch=batch,
+            device=device,
+        )
+
         # Step 6: Build Modality
         modality = Modality(
             enabled=True,
@@ -437,6 +492,7 @@ class FlexibleStrategy(TrainingStrategy):
             positions=positions,
             context=prompt_embeds,
             context_mask=prompt_attention_mask,
+            cond_channels=cond_channels,
         )
 
         return ModalityProcessingResult(
@@ -669,6 +725,150 @@ class FlexibleStrategy(TrainingStrategy):
 
         return combined_latents, combined_positions, combined_timesteps, combined_loss_mask, targets
 
+    def _apply_scail_conditions(
+        self,
+        modality_config: ModalityConfig,
+        modality_key: str,
+        noisy_latents: Tensor,
+        positions: Tensor,
+        timesteps: Tensor,
+        loss_mask: Tensor | None,
+        data: LatentData,
+        batch: dict[str, Any],
+        device: torch.device,
+    ) -> tuple[Tensor, Tensor, Tensor, Tensor | None, Tensor | None]:
+        """Apply SCAIL-2 driving concatenation then in-context mask channels (video only).
+        Driving tokens are prepended (cond-first, like reference) so the target stays at the tail for
+        loss slicing; the mask channels are then written onto those leading driving tokens (the noisy
+        target keeps a zero mask). Returns the possibly-extended sequence tensors plus ``cond_channels``
+        (``None`` when no mask condition is present).
+        """
+        if modality_key != "video":
+            return noisy_latents, positions, timesteps, loss_mask, None
+
+        driving_token_count = 0
+        for cond in modality_config.conditions:
+            if isinstance(cond, DrivingConditionConfig):
+                noisy_latents, positions, timesteps, loss_mask, driving_token_count = self._apply_driving_condition(
+                    noisy_latents=noisy_latents,
+                    positions=positions,
+                    timesteps=timesteps,
+                    loss_mask=loss_mask,
+                    target_width=data.width,
+                    batch=batch,
+                    config=cond,
+                )
+
+        cond_channels = None
+        for cond in modality_config.conditions:
+            if isinstance(cond, MaskChannelsConditionConfig):
+                cond_channels = self._build_mask_channels(
+                    config=cond,
+                    batch=batch,
+                    total_tokens=noisy_latents.shape[1],
+                    driving_token_count=driving_token_count,
+                    target_frames=data.num_frames,
+                    target_height=data.height,
+                    target_width=data.width,
+                    device=device,
+                )
+
+        return noisy_latents, positions, timesteps, loss_mask, cond_channels
+
+    def _apply_driving_condition(
+        self,
+        noisy_latents: Tensor,
+        positions: Tensor,
+        timesteps: Tensor,
+        loss_mask: Tensor | None,
+        target_width: int,
+        batch: dict[str, Any],
+        config: DrivingConditionConfig,
+    ) -> tuple[Tensor, Tensor, Tensor, Tensor | None, int]:
+        """Prepend SCAIL-2 driving latents with a RoPE width offset (ΔW).
+        Driving latents share the target's frame/height/width, so their time and height coordinates
+        already match the target (same ``_get_video_positions`` grid); only the width axis is shifted
+        by ``width_offset`` so the driving tokens stay spatially detached. Driving tokens are clean
+        (timestep=0) and excluded from loss. Returns the driving token count for mask placement.
+        The apply/skip decision is batch-wide (concatenation changes the sequence length) but drawn
+        from the torch RNG for reproducibility, mirroring ``_apply_reference_condition``.
+        """
+        if torch.rand((), device=noisy_latents.device).item() >= config.probability:
+            return noisy_latents, positions, timesteps, loss_mask, 0
+
+        cond = self._patchify_latent_data(batch[config.latents_dir], "video")
+        drv_latents = cond.latents
+        batch_size, drv_seq_len, _ = drv_latents.shape
+        device = drv_latents.device
+        dtype = drv_latents.dtype
+
+        drv_positions = self._get_video_positions(
+            num_frames=cond.num_frames,
+            height=cond.height,
+            width=cond.width,
+            batch_size=batch_size,
+            fps=cond.fps,
+            device=device,
+        ).clone()
+        width_offset = (
+            config.width_offset
+            if config.width_offset is not None
+            else float(target_width * VIDEO_SCALE_FACTORS.width)
+        )
+        drv_positions[:, 2, ...] = drv_positions[:, 2, ...] + width_offset
+
+        drv_timesteps = torch.zeros(batch_size, drv_seq_len, device=device, dtype=dtype)
+        drv_loss_mask = torch.zeros(batch_size, drv_seq_len, dtype=torch.bool, device=device)
+
+        combined_latents = torch.cat([drv_latents, noisy_latents], dim=1)
+        combined_positions = torch.cat([drv_positions, positions], dim=2)
+        combined_timesteps = torch.cat([drv_timesteps, timesteps], dim=1)
+        combined_loss_mask = torch.cat([drv_loss_mask, loss_mask], dim=1) if loss_mask is not None else None
+
+        return combined_latents, combined_positions, combined_timesteps, combined_loss_mask, drv_seq_len
+
+    def _build_mask_channels(
+        self,
+        config: MaskChannelsConditionConfig,
+        batch: dict[str, Any],
+        total_tokens: int,
+        driving_token_count: int,
+        target_frames: int,
+        target_height: int,
+        target_width: int,
+        device: torch.device,
+    ) -> Tensor:
+        """Encode semantic masks into per-token channels written onto the leading driving tokens.
+        The mask tensor at ``batch[mask_dir]["mask"]`` is expected as ``[B, K+1, F_pix, H_pix, W_pix]``
+        (channel 0 = environment switch, ``1..K`` = binding slots). It is encoded to
+        ``temporal_factor * (K+1)`` per-token channels (56 for K=6) via ``encode_mask_channels`` and
+        placed on the driving tokens; the target tokens keep an all-zero mask (paper-faithful).
+        """
+        masks = batch[config.mask_dir]["mask"].to(device=device)
+        encoded = encode_mask_channels(
+            masks=masks,
+            temporal_factor=VIDEO_SCALE_FACTORS.time,
+            height_lat=target_height,
+            width_lat=target_width,
+            frames_lat=target_frames,
+        )
+        tokens = self._video_patchifier.patchify(encoded)  # [B, N, C_mask]
+        batch_size, n_region, c_mask = tokens.shape
+
+        cond_channels = tokens.new_zeros(batch_size, total_tokens, c_mask)
+        if driving_token_count == 0:
+            # No driving tokens (e.g. driving skipped by its probability): fall back to writing the
+            # mask onto the leading target tokens so the channel width still matches the model.
+            cond_channels[:, :n_region] = tokens
+        else:
+            if n_region != driving_token_count:
+                raise ValueError(
+                    f"mask token count ({n_region}) must equal the driving token count "
+                    f"({driving_token_count}); driving and mask must describe the same grid."
+                )
+            cond_channels[:, :driving_token_count] = tokens
+        return cond_channels
+
     @staticmethod
     def _compute_modality_loss(pred: Tensor, targets: Tensor, loss_mask: Tensor) -> Tensor:
         """Compute per-element MSE loss for a single modality. Returns [B,]."""