Have a look at <a href=\"https://github.com/IBM/Flash-GMM\" rel=\"nofollow\">https://github.com/IBM/Flash-GMM</a></p>\n","updatedAt":"2026-06-12T08:38:42.228Z","author":{"_id":"644d6b7d16703fd67029b4e7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/644d6b7d16703fd67029b4e7/OldM8M7ewRN2gRtzXI2pi.jpeg","fullname":"Ohad Eytan","name":"ohadeytan","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.912114679813385},"editors":["ohadeytan"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/644d6b7d16703fd67029b4e7/OldM8M7ewRN2gRtzXI2pi.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.10896","authors":[{"_id":"6a2ade914957fcdd3aac03ef","name":"Gal Bloch","hidden":false},{"_id":"6a2ade914957fcdd3aac03f0","name":"Ariel Gera","hidden":false},{"_id":"6a2ade914957fcdd3aac03f1","name":"Matan Orbach","hidden":false},{"_id":"6a2ade914957fcdd3aac03f2","user":{"_id":"644d6b7d16703fd67029b4e7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/644d6b7d16703fd67029b4e7/OldM8M7ewRN2gRtzXI2pi.jpeg","isPro":false,"fullname":"Ohad Eytan","user":"ohadeytan","type":"user","name":"ohadeytan"},"name":"Ohad Eytan","status":"claimed_verified","statusLastChangedAt":"2026-06-12T07:11:53.710Z","hidden":false},{"_id":"6a2ade914957fcdd3aac03f3","name":"Assaf Toledo","hidden":false}],"publishedAt":"2026-06-09T00:00:00.000Z","submittedOnDailyAt":"2026-06-12T00:00:00.000Z","title":"Flash-GMM: A Memory-Efficient Kernel for Scalable Soft Clustering","submittedOnDailyBy":{"_id":"644d6b7d16703fd67029b4e7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/644d6b7d16703fd67029b4e7/OldM8M7ewRN2gRtzXI2pi.jpeg","isPro":false,"fullname":"Ohad Eytan","user":"ohadeytan","type":"user","name":"ohadeytan"},"summary":"We present Flash-GMM, a fused Triton kernel for efficient computation of Gaussian Mixture Models (GMMs) over large-scale data in a single GPU pass. By eliminating the need to materialize the full responsibility matrix in GPU memory, Flash-GMM achieves a 20times speedup over existing implementations and enables training on datasets more than 100times larger than previously feasible on one device. To demonstrate its impact, we integrate Flash-GMM into the IVF coarse quantizer for approximate nearest-neighbor (ANN) search. We show that soft GMM clustering is now a viable drop-in replacement for k-means, and that GMM responsibilities can be leveraged to assign border vectors to multiple clusters. Our approach reaches fixed recall targets with up to 1.7times fewer distance computations, or equivalently, yields +2--12 recall@10 at matched computational cost. We release the kernel as an open-source project.","upvotes":0,"discussionId":"6a2ade924957fcdd3aac03f4","githubRepo":"https://github.com/IBM/Flash-GMM","githubRepoAddedBy":"user","ai_summary":"Flash-GMM introduces an efficient fused Triton kernel for Gaussian Mixture Models that achieves significant speedup and enables processing much larger datasets on a single GPU.","ai_keywords":["Gaussian Mixture Models","Triton kernel","responsibility matrix","approximate nearest-neighbor search","k-means","soft clustering","IVF coarse quantizer","distance computations","recall@10"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":11,"organization":{"_id":"6760ab6c5c9a8ea8370ab95b","name":"ibm-research","fullname":"IBM Research","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/npxapKcW-cXX3J2JBl2vY.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"6760ab6c5c9a8ea8370ab95b","name":"ibm-research","fullname":"IBM Research","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/npxapKcW-cXX3J2JBl2vY.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.10896.md","query":{}}">
Flash-GMM: A Memory-Efficient Kernel for Scalable Soft Clustering
Abstract
Flash-GMM introduces an efficient fused Triton kernel for Gaussian Mixture Models that achieves significant speedup and enables processing much larger datasets on a single GPU.
We present Flash-GMM, a fused Triton kernel for efficient computation of Gaussian Mixture Models (GMMs) over large-scale data in a single GPU pass. By eliminating the need to materialize the full responsibility matrix in GPU memory, Flash-GMM achieves a 20times speedup over existing implementations and enables training on datasets more than 100times larger than previously feasible on one device. To demonstrate its impact, we integrate Flash-GMM into the IVF coarse quantizer for approximate nearest-neighbor (ANN) search. We show that soft GMM clustering is now a viable drop-in replacement for k-means, and that GMM responsibilities can be leveraged to assign border vectors to multiple clusters. Our approach reaches fixed recall targets with up to 1.7times fewer distance computations, or equivalently, yields +2--12 recall@10 at matched computational cost. We release the kernel as an open-source project.
Community
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.10896 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.10896 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.10896 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.