--- title: Generative modeling with sparse transformers | daily.dev description: OpenAI introduces the Sparse Transformer, a deep neural network that reformulates the self-attention mechanism to handle sequences up to 30x longer than... canonical: https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Generative modeling with sparse transformers | daily.dev og:description: OpenAI introduces the Sparse Transformer, a deep neural network that reformulates the self-attention mechanism to handle sequences up to 30x longer than... og:url: https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja og:image: https://api.daily.dev/og/posts/UfLMehyjA.png og:image:alt: Generative modeling with sparse transformers og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Generative modeling with sparse transformers **[OpenAI](https://daily.dev/sources/openai)** · 7 min read · 0 upvotes · 0 comments ## Summary OpenAI introduces the Sparse Transformer, a deep neural network that reformulates the self-attention mechanism to handle sequences up to 30x longer than previously feasible. By computing attention over sparse subsets of input positions rather than all positions, memory requirements drop dramatically — enabling 128-layer networks on images, text, and raw audio. Two factorized sparse attention patterns are proposed: strided (row/column) and fixed. The model achieves state-of-the-art density estimation on CIFAR-10, Enwik8, and ImageNet 64, and can generate ~5 seconds of raw audio at 12 kHz. Block-sparse GPU kernels and example code are open-sourced on GitHub. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [013\_the\_topological\_transformer\_training\_tauformer](https://daily.dev/posts/013-the-topological-transformer-training-tauformer-fttmkjdal) · Hacker News · 1 upvotes · 0 comments - [These startups are chasing the next big thing in LLMs](https://daily.dev/posts/these-startups-are-chasing-the-next-big-thing-in-llms-g8wfdvibc) · MIT Technology Review · 0 upvotes · 0 comments - [Fundamentals of AI: Inside the transformer](https://daily.dev/posts/fundamentals-of-ai-inside-the-transformer-szwloqlw1) · Cisco · 0 upvotes · 0 comments - [Beyond Quantization: Bringing Sparse Inference to PyTorch – PyTorch](https://daily.dev/posts/beyond-quantization-bringing-sparse-inference-to-pytorch-pytorch-hctyivuyg) · PyTorch · 0 upvotes · 0 comments --- Tags: [#deep-learning](https://daily.dev/tags/deep-learning), [#neural-networks](https://daily.dev/tags/neural-networks) [View this post on daily.dev](https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Generative modeling with sparse transformers","url":"https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja"},"datePublished":"2026-07-09T18:39:25.023Z","dateModified":"2026-07-09T19:39:50.561Z","description":"OpenAI introduces the Sparse Transformer, a deep neural network that reformulates the self-attention mechanism to handle sequences up to 30x longer than...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e5a6aa8c99087fc83fc25f9ba3237825?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e5a6aa8c99087fc83fc25f9ba3237825?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"OpenAI","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"OpenAI","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/984b3632316f4afaa3d5213d5c64fe65","url":"https://daily.dev/sources/openai"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/generative-modeling-with-sparse-transformers-uflmehyja","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"deep-learning,neural-networks","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"OpenAI","item":"https://daily.dev/sources/openai"},{"@type":"ListItem","position":3,"name":"Generative modeling with sparse transformers"}]} ```