# ChronoEdit Requirements (Merged) # Python 3.12+ # # This file merges requirements.txt and requirements2.txt with conflicts resolved. # Core ML/DL frameworks accelerate==1.8.1 diffusers==0.35.2 peft==0.17.1 megatron-core==0.12.1 peft==0.15.2 transformer-engine==1.13.0 transformers==4.57.1 ninja qwen-vl-utils==0.0.14 #flash-attn==2.6.3 # Data processing numpy==1.26.4 # Using 1.26.4 for stability (2.2.6 in req2.txt may have compatibility issues) pandas==2.2.3 pillow==11.1.0 opencv-python==4.11.0.86 opencv-python-headless==4.11.0.86 scikit-image==0.25.2 scipy==1.15.3 # Video/Media processing av==15.0.0 decord==0.6.0 imageio[pyav,ffmpeg]==2.37.0 imageio-ffmpeg==0.6.0 tifffile==2025.5.10 ffmpegcv # Configuration/Hydra hydra-core==1.3.2 omegaconf==2.3.0 pyyaml==6.0.2 # Tokenizers and text processing sentencepiece==0.2.0 tiktoken==0.9.0 tokenizers==0.22 ftfy==6.3.1 regex==2024.11.6 # Data loading and storage webdataset==0.2.111 zarr==2.18.3 tensorstore==0.1.45 numcodecs==0.13.1 # AWS/S3 support boto3==1.38.31 botocore==1.38.46 s3transfer==0.13.1 fsspec==2025.5.1 # Model storage and optimization safetensors==0.5.3 torchprofile==0.0.4 # Utilities einops==0.8.1 attrs==25.3.0 attr==0.3.2 asciitree==0.3.3 tyro==0.9.32 tqdm==4.67.1 psutil==7.0.0 loguru==0.7.3 rich==14.0.0 # Logging/Tracking wandb==0.21.0 # Infrastructure fvcore==0.1.5.post20221221 iopath==0.1.10 huggingface-hub==0.35.3 # Plotting/Visualization matplotlib==3.10.3 joblib==1.5.1 # Optional: Testing (keep if needed, remove if not running tests) coverage==7.9.2 pytest==8.4.1 pytest-cov==6.2.1 pytest-mock==3.14.1 pytest-random-order==1.2.0 pytest-xdist==3.8.0 ipdb==0.13.13 ipython==8.37.0 # Optional: Guardrails (keep if using content safety, remove otherwise) better-profanity==0.7.0 nltk==3.9.1 retinaface-py==0.0.2 # Standard dependencies (required by other packages) annotated-types==0.7.0 asttokens==3.0.0 certifi==2025.7.9 charset-normalizer==3.4.2 decorator==5.2.1 executing==2.2.0 filelock==3.18.0 idna==3.10 importlib-metadata==8.7.0 jedi==0.19.2 jinja2==3.1.6 jmespath==1.0.1 markupsafe==3.0.2 ninja==1.11.1.4 nvidia-ml-py==12.535.133 orjson==3.11.3 parso==0.8.4 pexpect==4.9.0 platformdirs==4.3.8 prompt-toolkit==3.0.51 protobuf==4.25.3 ptyprocess==0.7.0 pure-eval==0.2.3 pydantic==2.11.7 pydantic-core==2.33.2 pygments==2.19.2 pynvml==12.0.0 pyparsing==3.2.3 python-dateutil==2.9.0.post0 pytz==2025.2 requests==2.32.4 setuptools==80.9.0 six==1.17.0 stack-data==0.6.3 tomli==2.2.1 tomlkit==0.13.3 traitlets==5.14.3 typing-extensions==4.14.1 tzdata==2025.2 urllib3==2.5.0 wrapt==1.17.2 yacs==0.1.8 zipp==3.23.0
ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation,you can use this node in comfyUI,and Vram >8G
1.Installation
In the ./ComfyUI/custom_nodes directory, run the following:
git clone https://github.com/smthemex/ComfyUI_ChronoEdit_SM
2.requirements
pip install -r requirements.txt3.checkpoints
5 .wan T5 clipvison vae links
├── ComfyUI/models/gguf # or fp8
| ├── ChronoEdit-14B-Q6_K.gguf # or Q8
├── ComfyUI/models/vae
| ├──Wan2.1_VAE.pth
├── ComfyUI/models/clip
| ├──umt5_xxl_fp8_e4m3fn_scaled.safetensors
| ├──qwen_2.5_vl_7b.safetensors # OPTIONAL if use vl 可选 太慢
├── ComfyUI/models/clip_vision
| ├──clip_vision_h.safetensors
├── ComfyUI/models/loras
| ├──chronoedit_distill_lora.safetensors
| ├──upsample_lora_diffusers.safetensors
@article{wu2025chronoedit,
title={ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation},
author={Wu, Jay Zhangjie and Ren, Xuanchi and Shen, Tianchang and Cao, Tianshi and He, Kai and Lu, Yifan and Gao, Ruiyuan and Xie, Enze and Lan, Shiyi and Alvarez, Jose M. and Gao, Jun and Fidler, Sanja and Wang, Zian and Ling, Huan},
journal={arXiv preprint arXiv:2510.04290},
year={2025}
}