CPI-Bench: A Comprehensive,Practical and Intelligent Benchmark for Real-World Image Editing</p>\n","updatedAt":"2026-08-17T03:51:09.029Z","author":{"_id":"69e5c11e8605616ee4dcd92e","avatarUrl":"/avatars/edd17a0ed798a178df226e2ea25193a4.svg","fullname":"TaobaoTmall-AlgorithmProducts","name":"TaobaoTmall-AlgorithmProducts","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":7,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7344925999641418},"editors":["TaobaoTmall-AlgorithmProducts"],"editorAvatarUrls":["/avatars/edd17a0ed798a178df226e2ea25193a4.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.14546","authors":[{"_id":"6a828385b601d59c65281470","name":"Qinye Zhou","hidden":false},{"_id":"6a828385b601d59c65281471","name":"Jun Zheng","hidden":false},{"_id":"6a828385b601d59c65281472","name":"Yongchao Du","hidden":false},{"_id":"6a828385b601d59c65281473","name":"Yuan Wang","hidden":false},{"_id":"6a828385b601d59c65281474","name":"Zhengrui Chen","hidden":false},{"_id":"6a828385b601d59c65281475","name":"Zuan Gao","hidden":false},{"_id":"6a828385b601d59c65281476","name":"Taihang Hu","hidden":false},{"_id":"6a828385b601d59c65281477","name":"Chao Lin","hidden":false},{"_id":"6a828385b601d59c65281478","name":"Yefeng Shen","hidden":false},{"_id":"6a828385b601d59c65281479","name":"Xingjian Wang","hidden":false},{"_id":"6a828385b601d59c6528147a","name":"Zhao Wang","hidden":false},{"_id":"6a828385b601d59c6528147b","name":"Zhengtao Wu","hidden":false},{"_id":"6a828385b601d59c6528147c","name":"Xiaoli Xu","hidden":false},{"_id":"6a828385b601d59c6528147d","name":"Zhengze Xu","hidden":false},{"_id":"6a828385b601d59c6528147e","name":"Hao Yan","hidden":false},{"_id":"6a828385b601d59c6528147f","name":"Denghui Yang","hidden":false},{"_id":"6a828385b601d59c65281480","name":"Yuhang Yu","hidden":false},{"_id":"6a828385b601d59c65281481","name":"Huayu Zhang","hidden":false},{"_id":"6a828385b601d59c65281482","name":"Mingzhou Zhang","hidden":false},{"_id":"6a828385b601d59c65281483","name":"Mengting Chen","hidden":false}],"publishedAt":"2026-08-14T00:00:00.000Z","submittedOnDailyAt":"2026-08-17T00:00:00.000Z","title":"CPI-Bench: A Comprehensive,Practical and Intelligent Benchmark for Real-World Image Editing","submittedOnDailyBy":{"_id":"69e5c11e8605616ee4dcd92e","avatarUrl":"/avatars/edd17a0ed798a178df226e2ea25193a4.svg","isPro":true,"fullname":"TaobaoTmall-AlgorithmProducts","user":"TaobaoTmall-AlgorithmProducts","type":"user","name":"TaobaoTmall-AlgorithmProducts"},"summary":"With the rapid advancement of image editing models and their widespread application across various domains, there is an increasingly urgent need to deploy these model capabilities directly into real-world scenarios. However, existing benchmarks remain confined to simple single-image tasks, suffering from limited coverage dimensions and an inability to effectively differentiate performance among diverse models. Consequently, they fail to reliably evaluate model performance in complex multi-image editing, highly demanding reasoning instructions, and practical deployment settings. To address these limitations, we propose CPI-Bench, a Comprehensive, Practical andIntelligent benchmark for real-world image editing. CPI-Bench comprises three core subsets: CPI-General-Bench, which comprehensively covers diverse editing tasks and pioneers the inclusion of multi-image editing evaluation; CPI-Practical-Bench, which focuses on high-frequency real-user application scenarios; and CPI-Intelligent-Bench, which is dedicated to evaluating capabilities in highly demanding reasoning-based editing. Evaluation results of mainstream image editing models based on CPI-Bench demonstrate that CPI-Bench enhances performance differentiation among models. It provides a comprehensive and reliable quantification of gaps in general editing capabilities, practical deployment efficacy, and advanced reasoning-based editing, offering invaluable guidance for the future optimization of image editing models. Crucially, our ranking analysis reveals that CPI-Bench achieves the highest alignment with the Arena Image Edit Leaderboard, indicating it faithfully captures the preferences and perceptual judgments of human evaluators, serving as a robust proxy for real-world user experience.","upvotes":10,"discussionId":"6a828385b601d59c65281484","projectPage":"https://huggingface.co/datasets/TaobaoTmall-AlgorithmProducts/CPI-benchmark","githubRepo":"https://github.com/zqyzzz/CPI-benchmark","githubRepoAddedBy":"user","ai_summary":"CPI-Bench is a comprehensive benchmark for real-world image editing that evaluates multi-image tasks, practical applications, and reasoning-based editing to better differentiate model performance.","ai_keywords":["CPI-Bench","multi-image editing","reasoning-based editing","image editing models","Arena Image Edit Leaderboard"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":2,"organization":{"_id":"68be41370a3fcebdcad6516a","name":"alibabagroup","fullname":"alibaba","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68be3ab7e52df040b2cf80dc/li4G29u_EGswyTN1Sm_Kq.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6661ad9bc229e1da8e9b939d","avatarUrl":"/avatars/45d83ebdc1a2fc027309eb28f2584ef2.svg","isPro":false,"fullname":"chen","user":"mathildachen","type":"user"},{"_id":"65a909fe8581aad8c97a67d3","avatarUrl":"/avatars/96570e47117e957543d9f0fe5e1d9d57.svg","isPro":false,"fullname":"liutao","user":"byliutao","type":"user"},{"_id":"6470125bd742e9ef65222017","avatarUrl":"/avatars/6f1bfd673729f43f2a2a92e13c0287b8.svg","isPro":false,"fullname":"Zhengrui Chen","user":"zhengrchan","type":"user"},{"_id":"6381847a471a4550ff298c63","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6381847a471a4550ff298c63/RTKepvX67R6pLiiUidpUO.png","isPro":false,"fullname":"Jun","user":"zxbsmk","type":"user"},{"_id":"67594a15e561be24b25697c9","avatarUrl":"/avatars/73c8ab719c710903c10fd887ee0fcaca.svg","isPro":false,"fullname":"huayuzhang","user":"rainzhang666","type":"user"},{"_id":"634bec6ca5d4e109c6aa4895","avatarUrl":"/avatars/2f55327097b3782f6f4f59041f3b04b5.svg","isPro":false,"fullname":"hu","user":"taihang","type":"user"},{"_id":"650bbd1487dcda6616acb1e2","avatarUrl":"/avatars/925b83b15b239307b2cd444b8006a2ae.svg","isPro":false,"fullname":"lmnhsp","user":"lmnhsp","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"647d92e09bb822b5cd3c7692","avatarUrl":"/avatars/73de68a74630667ea73e8dc86acd88ae.svg","isPro":false,"fullname":"Zhengtao Wu","user":"wuzht","type":"user"},{"_id":"69e5c11e8605616ee4dcd92e","avatarUrl":"/avatars/edd17a0ed798a178df226e2ea25193a4.svg","isPro":true,"fullname":"TaobaoTmall-AlgorithmProducts","user":"TaobaoTmall-AlgorithmProducts","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"68be41370a3fcebdcad6516a","name":"alibabagroup","fullname":"alibaba","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68be3ab7e52df040b2cf80dc/li4G29u_EGswyTN1Sm_Kq.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.14546.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2608.14546 in a model README.md to link it from this page.

Cite arxiv.org/abs/2608.14546 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2608.14546 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.