--- title: Why Is Everyone Hyped for Qwen 4? (Architecture Explained) description: Alibaba's Qwen team released Qwen3.8 Flash Next, an experimental preview model with a technical report described as an early look at the architecture planned... canonical: https://daily.dev/posts/ykcli5opt twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Why Is Everyone Hyped for Qwen 4? (Architecture Explained) | daily.dev og:description: Alibaba's Qwen team released Qwen3.8 Flash Next, an experimental preview model with a technical report described as an early look at the architecture planned... og:url: https://daily.dev/posts/ykcli5opt og:image: https://api.daily.dev/og/posts/ykclI5opt.png og:image:alt: Post cover image 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. # Why Is Everyone Hyped for Qwen 4? (Architecture Explained) **[Qwen](https://daily.dev/sources/qwen)** · [@nerdalytics](https://daily.dev/nerdalytics) · 19 upvotes · 0 comments ## Summary Alibaba's Qwen team released Qwen3.8 Flash Next, an experimental preview model with a technical report described as an early look at the architecture planned for Qwen 4. The design mixes a large mixture-of-experts network (125B total parameters, ~6B active per token) with hybrid attention: most layers use a compact linear-memory mechanism (Gated DeltaNet) while every fourth layer uses sparse attention for more precise lookups, plus residual gating to preserve signal across layers, an n-gram lookup table stored in system RAM for extra capacity, and a Muon optimizer paired with AdamW for stable, cheaper training. In Alibaba's own benchmarks, the preview model beat its larger predecessor on 8 of 14 benchmarks using roughly one-ninth the training compute, and its sparse attention kernel showed up to 7.6x faster prompt processing in component tests. The architecture is not final, memory requirements remain high despite low active-parameter counts, and the preview license restricts certain commercial uses. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: --- Tags: [#qwen](https://daily.dev/tags/qwen), [#mixture-of-experts](https://daily.dev/tags/mixture-of-experts), [#alibaba](https://daily.dev/tags/alibaba) [View this post on daily.dev](https://daily.dev/posts/ykcli5opt) ```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":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/ykcli5opt","headline":"Why Is Everyone Hyped for Qwen 4? (Architecture Explained)","text":"Shared: Why Is Everyone Hyped for Qwen 4? (Architecture Explained)","url":"https://daily.dev/posts/ykcli5opt","datePublished":"2026-10-19T04:50:41.198Z","dateModified":"2026-10-19T04:51:33.496Z","author":{"@type":"Person","name":"nerdalytics","url":"https://daily.dev/nerdalytics","image":"https://avatars.githubusercontent.com/u/97166791?v=4","description":"Husband. Cat and dog dad. Nerd.","interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":7230}},"image":"https://media.daily.dev/image/upload/s--ZrL_HSsR--/f_auto/v1722860399/public/Placeholder%2006","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":19},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"sharedContent":{"@type":"WebPage","url":"https://api.daily.dev/r/JKbv4aVCA"},"isPartOf":{"@type":"WebPage","url":"https://daily.dev/squads/qwen","name":"Qwen"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Qwen","item":"https://daily.dev/squads/qwen"},{"@type":"ListItem","position":3,"name":"Why Is Everyone Hyped for Qwen 4? (Architecture Explained)"}]} ```