[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85227-en":3,"doc-seo-85227-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},85227,13056703019662,"Evangeline","https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188",8,"Research & Report","Navigating the Open-Source Model Ecosystem: An Empirical Study of Creator Practices in Artistic Image Generation","Open-sourcing of powerful image generation models has enabled an active ecosystem where creators curate, combine, and repurpose community-contributed models, unlike closed-source workflows. This paper introduces the first large-scale empirical study of creator model-usage behavior in the open-source artistic image generation ecosystem. A new dataset of 6 million images includes embedded generation metadata—prompts and models—linked to 22.4K base models and 154K LoRA models. Results highlight ecosystem strengths and obstacles, and the dataset is publicly released to support creators and researchers.","Navigating the Open-Source Model Ecosystem: An Empirical Study of Creator Practices in Artistic Image Generation  \nYiluo Wei  \nThe Hong Kong University of Science and Technology (Guangzhou) Guangzhou, China  \nYupeng He  \nThe Hong Kong University of Science and Technology (Guangzhou) Guangzhou, China  \nQiming Ye  \nThe Hong Kong University of Science and Technology (Guangzhou) Guangzhou, China  \nGareth Tyson  \nThe Hong Kong University of Science and Technology (Guangzhou) Guangzhou, China  \narXiv :2607 . 10538v1 [ cs .HC] 12 Jul 2026  \nAbstract  \nThe open-sourcing of powerful image generation models has created a vibrant ecosystem where creators curate and combine a vast array of community-contributed models. This practice stands in sharp contrast to using closed-source tools like Midjourney. Yet, little is known about these emerging creative workflows. To bridge this gap, this paper presents the first large-scale empirical study of creator model usage behavior within this open-source image generation ecosystem. We construct a novel dataset of 6 million images with their embedded generation metadata—a detailed recipe of the creation process, including the models used and the prompts. By linking the usage of 22.4K base models and 154K LoRA models to the images, our findings underscore the ecosystem’s unique strengthsand its inherent obstacles. This provides valuable insights for making this ecosystem more sustainable and innovative. Moreover, we make our dataset publicly available,1 providing creators with practical references for producing better artworks and researchers to facilitate further studies.  \nCCS Concepts  \n• Human-centered computing → Empirical studies in collaborative and social computing.  \nKeywords  \nPixiv, Creator, Artwork, Generative AI  \n1 Introduction  \nThe commodification of AI Generated Content (AIGC) has reshaped online creative communities. The advent of powerful generative diffusion models [36] has been a key catalyst in this transformation [6] . The open-sourcing of highly capable models like Stable Diffusion [31] has further democratized access to this technology, allowing users not only to generate images but also to easily fine-tune and extend models for specific styles and concepts [10, 32, 45] .  \nThis accessibility has fueled the rapid emergence of a vibrant open-source generative AI ecosystem [7]. Platforms like Civitai [41] now host hundreds of thousands of community-contributed models. In this new paradigm, different from the close-source commercial platforms such as Midjourney [25, 43], creators are no longer passive users of a single tool. Instead, they are active curators who can select from, combine, and even develop a diverse array of models to realize their artistic visions [21] . For example, a common practice is to combine a foundational “base” model with lightweight LoRA (Low-Rank Adaptation) finetuning [15] for nuanced adjustments.  \n1[https://huggingface.co/datasets/Wei-Yiluo/Pixiv-AI-generation-metadata](https://huggingface.co/datasets/Wei-Yiluo/Pixiv-AI-generation-metadata)  \nThis complex interplay of tools, techniques, and community-driven development represents a fundamental shift in digital creation.  \nHowever, while the technical capabilities and the proliferation of models have been well documented [1, 4], a critical gap remains in our understanding of the creative practices that have emerged within this ecosystem — specifically, how this massive and diverse array of models are selected, used, and combined by creators. To fully appreciate the value of the open-source paradigm, it is crucial to systematically investigate the key differences in its highly modular creative workflows compared to close-source alternatives. Our goal is to identify its unique strengths and inherent challenges. Such an understanding will directly benefit creators seeking to master new workflows to empower their artworks, developers aiming to build more useful models and tools, and platforms striving to","cbCaifPldEZA5Sxz","https://ap.wps.com/l/cbCaifPldEZA5Sxz","pdf",4735803,2,1,16,"English","en",105,"# Introduction\n## Background and motivation\n## Key research gap\n## Study approach and dataset construction","[{\"question\":\"What is the main goal of the study on the open-source image generation ecosystem?\",\"answer\":\"The study aims to understand how creators select, use, and combine a large and diverse set of models, identifying both unique strengths and inherent challenges of the modular open-source workflow.\"},{\"question\":\"How was the dataset constructed and what does it contain?\",\"answer\":\"The authors build a large-scale dataset by collecting AI-tagged images on Pixiv and extracting generation metadata embedded in image file headers, capturing prompts, parameters, and the specific models used.\"},{\"question\":\"Which model types and scale are analyzed in the findings?\",\"answer\":\"Findings link usage patterns across 22.4K base models and 154K LoRA models to a dataset of 6 million images, then relate these configurations to engagement signals such as views and likes.\"}]",1784201862,40,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"navigating-the-open-source-model-ecosystem-an-empirical-study-of-creator-practices-in-artistic-image-generation","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/navigating-the-open-source-model-ecosystem-an-empirical-study-of-creator-practices-in-artistic-image-generation/85227/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the main goal of the study on the open-source image generation ecosystem?","Question",{"text":75,"@type":76},"The study aims to understand how creators select, use, and combine a large and diverse set of models, identifying both unique strengths and inherent challenges of the modular open-source workflow.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How was the dataset constructed and what does it contain?",{"text":80,"@type":76},"The authors build a large-scale dataset by collecting AI-tagged images on Pixiv and extracting generation metadata embedded in image file headers, capturing prompts, parameters, and the specific models used.",{"name":82,"@type":73,"acceptedAnswer":83},"Which model types and scale are analyzed in the findings?",{"text":84,"@type":76},"Findings link usage patterns across 22.4K base models and 154K LoRA models to a dataset of 6 million images, then relate these configurations to engagement signals such as views and likes.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":29,"slug":118},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]