A high-performing, fully open-source 3B-parameter text-to-image diffusion model with a systematic training recipe.</p>\n","updatedAt":"2026-06-11T02:36:00.055Z","author":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","fullname":"taesiri","name":"taesiri","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":314,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7940531969070435},"editors":["taesiri"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.11289","authors":[{"_id":"6a2a190780a9c7c6830c0e8e","user":{"_id":"6742a0adf6b05a786e6cc6e6","avatarUrl":"/avatars/4d45558677bd2045588aeb4125733073.svg","isPro":false,"fullname":"Boya Zeng","user":"boyazeng","type":"user","name":"boyazeng"},"name":"Boya Zeng","status":"claimed_verified","statusLastChangedAt":"2026-06-11T08:38:52.675Z","hidden":false},{"_id":"6a2a190780a9c7c6830c0e8f","name":"Tianze Luo","hidden":false},{"_id":"6a2a190780a9c7c6830c0e90","name":"Shu Pu","hidden":false},{"_id":"6a2a190780a9c7c6830c0e91","name":"Jucheng Shen","hidden":false},{"_id":"6a2a190780a9c7c6830c0e92","name":"Taiming Lu","hidden":false},{"_id":"6a2a190780a9c7c6830c0e93","name":"Gabriel Sarch","hidden":false},{"_id":"6a2a190780a9c7c6830c0e94","name":"Zhuang Liu","hidden":false}],"publishedAt":"2026-06-09T00:00:00.000Z","submittedOnDailyAt":"2026-06-11T00:00:00.000Z","title":"i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user","name":"taesiri"},"summary":"Diffusion models have consistently driven progress in text-to-image generation. However, it is challenging to attribute recent progress to specific modeling and data choices: state-of-the-art open-weight models provide limited ablations, and do not disclose their training data and full training details. The research community needs fully open (weights, data, and code) models as a foundation for further research; yet existing fully open models still fall significantly short of leading models in performance. In this project, we conduct a systematic investigation of the modeling and data design choices in text-to-image diffusion training and inference with 300+ controlled experiments totaling 700K+ TPU v6e hours. Our experiments highlight several empirical findings (e.g., equal weighting is a strong default for mixing curated datasets) and simple design decisions (e.g., larger text encoder adapters improve performance with minimal added parameters) for training strong models. Guided by these insights, we train i1, a 3B-parameter text-to-image diffusion model using only publicly available datasets. i1 is competitive with leading models on five representative benchmarks (GenEval, DPG, PRISM, CVTG-2K, and LongText), and outperforms the best existing fully open model by 29.5 absolute percentage points on average. We provide the i1 checkpoints, training and inference code, and the data processing pipeline. Together, our findings and the i1 recipe establish a practical foundation for future open research in text-to-image diffusion models. Our code is available at https://github.com/zlab-princeton/i1.","upvotes":5,"discussionId":"6a2a190880a9c7c6830c0e95","projectPage":"https://zlab-princeton.github.io/i1/","githubRepo":"https://github.com/zlab-princeton/i1","githubRepoAddedBy":"user","ai_summary":"A comprehensive experimental study of text-to-image diffusion models reveals key design choices and training insights leading to the development of i1, a 3B-parameter model that matches leading performance while maintaining full openness.","ai_keywords":["text-to-image diffusion models","controlled experiments","publicly available datasets","model training","model inference","GenEval","DPG","PRISM","CVTG-2K","LongText"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":38},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6825a649f7587837c603ffd6","avatarUrl":"/avatars/93e6adb1234d3e873957961a3a3a0d28.svg","isPro":true,"fullname":"bigpie","user":"nacjadcans","type":"user"},{"_id":"6589b61dbfdd9f4410af9b7d","avatarUrl":"/avatars/7d644e750f6084d4f24e332135cc5be8.svg","isPro":false,"fullname":"Hu","user":"Irving1","type":"user"},{"_id":"6a2ae6c2e36bc84d91b6e7cc","avatarUrl":"/avatars/abf4b4c0020f9332b6827952cc53163e.svg","isPro":false,"fullname":"mmgood","user":"mmgood","type":"user"},{"_id":"649be88f867d442094248239","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/NFwa74mbWNEBfHlt82hpp.jpeg","isPro":false,"fullname":"SAMBIT CHAKRABORTY","user":"sambitchakhf03","type":"user"},{"_id":"656a9b9f9496f21be8271f1b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/656a9b9f9496f21be8271f1b/oOcpLo9-anHJJ6kJ0BF1P.jpeg","isPro":true,"fullname":"TaiMing","user":"TaiMingLu","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.11289.md"}">

Get this paper in your agent:

Don't have the latest CLI?