You need to agree to share your Hugging Face information to access this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Access to these weights is granted manually by the authors. Please tell us your name, affiliation and intended use.
Log in or Sign Up to review the conditions and access this model content.
videoeditor-14b-long
Code: https://github.com/massyzs/Video-Editor-MoE (scripts/moe/videoeditor-14b/)
Optimizer state: removed (weights only).
Long-video variant of videoeditor-14b: the released step-18,630 weights further trained on the same data re-cut to 121 or 81 frames per sample (frame count chosen per video, batches homogeneous in frame count) for half an epoch (37,260 steps, 192 samples per update).
Training checkpoint: step 37,260 (end of the long-video pass). Benchmark scores for the long-video variants are not yet available.
Package contents
The weights are shipped as a tar.gz stream split into ~20 GB chunks (Hugging Face limits single files to 50 GB).
| chunk | size | sha256 |
|---|---|---|
videoeditor-14b-long.tar.gz.part-00 |
21.47 GB | 404e0339e7afe7159c150f5a799b13068b50af297aa72f279e4e257ad1651133 |
videoeditor-14b-long.tar.gz.part-01 |
0.38 GB | 37fff61db43015cd6664069b30492de3ec9881f4617c44ce9328373d82032fb8 |
Total: 2 chunks, 21.86 GB compressed / 27.97 GB extracted.
Reassemble and verify:
huggingface-cli download Massyzs/videoeditor-14b-long --local-dir videoeditor-14b-long-package # after your access request is approved
cd videoeditor-14b-long-package
cat videoeditor-14b-long.tar.gz.part-* | tar -xzf - -C <BASE_DIR>/ckpt/ # creates <BASE_DIR>/ckpt/videoeditor-14b-long/
cp SHA256SUMS.contents <BASE_DIR>/ckpt/ && (cd <BASE_DIR>/ckpt && sha256sum -c SHA256SUMS.contents) # optional check
MANIFEST.json lists every payload file with its size and sha256, and every chunk with its sha256 (SHA256SUMS.parts).
Extracted files:
| file | size | sha256 |
|---|---|---|
videoeditor-14b-long/meta.json |
455 B | 4136c86a7b973c00d8719d824023a70a62cf66cfdb5937f577688037fa38e792 |
videoeditor-14b-long/moe_v3_1_trainable.safetensors |
27.97 GB | 4e30e46510ad29fb41dfbc4be92c8900f15643c70bbdc5bb60b6b7573b9c16e1 |
Base weights you also need
The package contains the full DiT weights (28.0 GB bf16); only the MLLM encoder (qwen/), the VAE and moe_expert_init.safetensors are needed in addition.
All frozen base components are published, already converted to the layout this code expects, in
Massyzs/kiwi-edit-5b-instruct-only-videoxfun:
huggingface-cli download Massyzs/kiwi-edit-5b-instruct-only-videoxfun --local-dir <BASE_DIR>/ckpt/kiwi-edit-instruct-only-videoxfun
See BASE_WEIGHTS.md for where each component comes from (Kiwi-Edit, Wan2.2) and its checksum.
Inference
python scripts/moe/videoeditor-14b/moe_v3_1_infer.py --base_dir <BASE_DIR> --ckpt <BASE_DIR>/ckpt/videoeditor-14b-long --num_frames 121 \
--src_video input.mp4 --prompt "Replace the red car with a blue truck" --name edited --out_dir <OUT_DIR>
# multi-step instruction: --prompts "Remove the dog ||| Convert to oil painting style"
Single 80 GB GPU. Use --num_frames 121 (or 81) for long inputs; shorter clips still work with the default 49. Output: <OUT_DIR>/<name>.mp4 plus lossless PNG frames in <OUT_DIR>/frames/<name>/.
Setup and usage (installation, where to put the downloaded weights, inputs and outputs) are described in the repository README.
Training
The training launcher is scripts/moe/videoeditor-14b/train_multinode.sh in the code repository: edit the configuration block at the top
(BASE_DIR, data paths, node addresses) and run bash scripts/moe/videoeditor-14b/train_multinode.sh <node rank> on every node.
scripts/moe/videoeditor-14b/train_smoke.sh is a single-node smoke test. The data formats are described in docs/DATA.md of the repository.
License
The released weights and code are under Apache-2.0. The frozen base components (Kiwi-Edit MLLM encoder / DiT base, Wan2.2 VAE) are not redistributed here and remain subject to their own licenses.
- Downloads last month
- -
Model tree for Massyzs/videoeditor-14b-long
Base model
linyq/kiwi-edit-5b-instruct-only-diffusers