{"id":18066295,"url":"https://github.com/Vchitect/FasterCache","last_synced_at":"2025-03-28T10:33:17.069Z","repository":{"id":259814285,"uuid":"878296348","full_name":"Vchitect/FasterCache","owner":"Vchitect","description":"[ICLR 2025] FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality","archived":false,"fork":false,"pushed_at":"2024-12-27T08:57:19.000Z","size":81307,"stargazers_count":201,"open_issues_count":12,"forks_count":10,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-03-21T11:07:17.707Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Vchitect.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-10-25T06:07:48.000Z","updated_at":"2025-03-21T02:59:14.000Z","dependencies_parsed_at":"2024-12-28T22:12:24.747Z","dependency_job_id":"9d785c12-e125-47ec-99e0-55c06ec7d571","html_url":"https://github.com/Vchitect/FasterCache","commit_stats":null,"previous_names":["vchitect/fastercache"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vchitect%2FFasterCache","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vchitect%2FFasterCache/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vchitect%2FFasterCache/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Vchitect%2FFasterCache/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Vchitect","download_url":"https://codeload.github.com/Vchitect/FasterCache/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246012797,"owners_count":20709515,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-10-31T07:01:18.676Z","updated_at":"2025-03-28T10:33:17.061Z","avatar_url":"https://github.com/Vchitect.png","language":"Python","funding_links":[],"categories":["Accelerate"],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\u003ch1\u003eFasterCache: Training-Free Video Diffusion Model Acceleration with High Quality\u003c/h1\u003e\u003c/div\u003e\n\n\n\n\u003cdiv align=\"center\"\u003e\n    \u003ca href=\"https://scholar.google.com/citations?user=FkkaUgwAAAAJ\u0026hl=en\" target=\"_blank\"\u003eZhengyao Lv\u003c/a\u003e\u003csup\u003e1\u003c/sup\u003e |\n    \u003ca href=\"https://chenyangsi.github.io/\" target=\"_blank\"\u003eChenyang Si\u003c/a\u003e\u003csup\u003e2‡\u003c/sup\u003e |\n    \u003ca href=\"\" target=\"_blank\"\u003eJunhao Song\u003c/a\u003e\u003csup\u003e3\u003c/sup\u003e |\n    \u003ca href=\"\" target=\"_blank\"\u003eZhenyu Yang\u003c/a\u003e\u003csup\u003e3\u003c/sup\u003e |\n    \u003ca href=\"https://mmlab.siat.ac.cn/yuqiao\" target=\"_blank\"\u003eYu Qiao\u003c/a\u003e\u003csup\u003e3\u003c/sup\u003e |\n    \u003ca href=\"https://liuziwei7.github.io/\" target=\"_blank\"\u003eZiwei Liu\u003c/a\u003e\u003csup\u003e2†\u003c/sup\u003e    |\n    \u003ca href=\"https://i.cs.hku.hk/~kykwong/\" target=\"_blank\"\u003eKwan-Yee K. Wong\u003c/a\u003e\u003csup\u003e1†\u003c/sup\u003e\n\u003c/div\u003e\n\u003cdiv align=\"center\"\u003e\n    \u003csup\u003e1\u003c/sup\u003eThe University of Hong Kong \u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; \n    \u003csup\u003e2\u003c/sup\u003eS-Lab, Nanyang Technological University \u003cbr\u003e\n    \u003csup\u003e3\u003c/sup\u003eShanghai Artificial Intelligence Laboratory\n\u003c/div\u003e\n\u003cdiv align=\"center\"\u003e(‡: Project lead; †: Corresponding authors)\u003c/div\u003e\n\n\u003cp align=\"center\"\u003e\n    \u003ca href=\"https://arxiv.org/abs/2410.19355\"\u003ePaper\u003c/a\u003e | \n    \u003ca href=\"https://vchitect.github.io/FasterCache/\"\u003eProject Page\u003c/a\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n    \u003ca href=\"https://hits.seeyoufarm.com\"\u003e\u003cimg src=\"https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Fgithub.com%2FVchitect%2FFasterCache\u0026count_bg=%2379C83D\u0026title_bg=%23555555\u0026icon=\u0026icon_color=%23E7E7E7\u0026title=Github+visitors\u0026edge_flat=false\"/\u003e\u003c/a\u003e\n    \u003ca href=\"https://hits.seeyoufarm.com\"\u003e\u003cimg src=\"https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Fvchitect.github.io%2FFasterCache%2F\u0026count_bg=%23C83D5D\u0026title_bg=%23555555\u0026icon=\u0026icon_color=%23E7E7E7\u0026title=Pages+visitors\u0026edge_flat=false\"/\u003e\u003c/a\u003e\n\u003c/p\u003e\n\n\n\n## About\n\nWe present ***FasterCache***, a novel training-free strategy designed to accelerate the inference of video diffusion models with high-quality generation. For more details and visual results, go checkout our [Project Page](https://vchitect.github.io/FasterCache/).\n\nhttps://github.com/user-attachments/assets/035c50c2-7b74-4755-ac1e-e5aa1cffba2a\n\n## News\n\n* (🔥 New) 2024/11/8 We support the multi-device inference script for CogvideoX\n* (🔥 New) 2024/11/8 We implemented FasterCache based on the Mochi\n\n## Usage\n\n### Installation\n\nRun the following instructions to create an Anaconda environment.\n\n```\nconda create -n fastercache python=3.10 -y\nconda activate fastercache\ngit clone https://github.com/Vchitect/FasterCache\ncd FasterCache\npip install -e .\n```\n\n### Inference\n\nWe currently support [Open-Sora 1.2](https://github.com/hpcaitech/Open-Sora), [Open-Sora-Plan 1.1](https://github.com/PKU-YuanGroup/Open-Sora-Plan), [Latte](https://github.com/Vchitect/Latte), [CogvideoX-2B\u00265B](https://github.com/THUDM/CogVideo),  [Vchitect 2.0](https://github.com/Vchitect/Vchitect-2.0) and [Mochi](https://github.com/genmoai/models). You can achieve accelerated sampling by executing the scripts we provide.\n\n- **Open-Sora**\n\n    For single-GPU inference on Open-Sora, run the following command:\n    ```\n    bash scripts/opensora/fastercache_sample_opensora.sh\n    ```\n\n    For multi-GPU inference on Open-Sora, run the following command:\n\n    ```\n    bash scripts/opensora/fastercache_sample_multi_device_opensora.sh\n    ```\n\n- **Open-Sora-Plan**\n\n    For single-GPU inference on Open-Sora-Plan, run the following command:\n    ```\n    bash scripts/opensora_plan/fastercache_sample_opensoraplan.sh\n    ```\n    \n    For multi-GPU inference on Open-Sora-Plan, run the following command:\n    \n    ```\n    bash scripts/opensora_plan/fastercache_sample_multi_device_opensoraplan.sh\n    ```\n\n- **Latte**\n\n\n    For single-GPU inference on Latte, run the following command:\n    ```\n    bash scripts/latte/fastercache_sample_latte.sh\n    ```\n    \n    For multi-GPU inference on Latte, run the following command:\n    \n    ```\n    bash scripts/latte/fastercache_sample_multi_device_latte.sh\n    ```\n\n- **CogVideoX**\n\n    For single-GPU or multi-GPU batched inference on CogVideoX-2B, run the following command:\n    ```\n    bash scripts/cogvideox/fastercache_sample_cogvideox.sh\n    ```\n\n    For multi-GPU inference on CogVideoX-2B, run the following command:\n    ```\n    bash scripts/cogvideox/fastercache_sample_cogvideox_multi_device.sh\n    ```\n\n    For inference on CogVideoX-5B, run the following command:\n\n    ```\n    bash scripts/cogvideox/fastercache_sample_cogvideox5b.sh\n    ```\n\n- **Vchitect 2.0**\n\n    For inference on Vchitect 2.0, run the following command:\n    ```\n    bash scripts/vchitect/fastercache_sample_vchitect.sh\n    ```\n\n* **Mochi**\n\n  We also provide acceleration scripts for Mochi. Before running these scripts, please follow the [official Mochi repository](https://github.com/genmoai/models) to complete model downloads, environment setup, and installation of genmo. Then, execute the following script:\n\n  ```\n  bash scripts/mochi/fastercache_sample_mochi.sh \n  ```\n\n## BibTeX\n\n```\n@inproceedings{lv2024fastercache,\n  title={FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality},\n  author={Lv, Zhengyao and Si, Chenyang and Song, Junhao and Yang, Zhenyu and Qiao, Yu and Liu, Ziwei and Kwan-Yee K. Wong},\n  booktitle={arxiv},\n  year={2024}\n}\n```\n\n## Acknowledgement\n\nThis repository borrows code from [VideoSys](https://github.com/NUS-HPC-AI-Lab/VideoSys), [Vchitect-2.0](https://github.com/Vchitect/Vchitect-2.0), [Mochi](https://github.com/genmoai/models), and [CogVideo](https://github.com/THUDM/CogVideo),.Thanks for their contributions!\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FVchitect%2FFasterCache","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FVchitect%2FFasterCache","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FVchitect%2FFasterCache/lists"}