[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-bend-language-aims-to-make-gpu-parallel-code-automatic":10,"sections":35},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},6739,"bend-language-aims-to-make-gpu-parallel-code-automatic","Bend Language Aims to Make GPU Parallel Code Automatic","Bend compiles ordinary code to run in parallel across CPUs and GPUs automatically, without hand-written threads or locks.","Bend wants you to write ordinary-looking code and have it run in parallel on a CPU or GPU, with no threads, locks, or mutexes required.\n\nBend is an open-source, high-level language built by the HigherOrderCO team that compiles down to HVM2, a runtime based on interaction combinators rather than a traditional stack machine. The pitch is straightforward: write recursive, Python-or-Haskell-flavored code with ordinary objects and closures, and the runtime finds the parallelism on its own, spreading work across CPU cores or CUDA-capable GPUs without the programmer writing any explicit thread-management code. The project resurfaced on Hacker News this week, pulling in 92 points and 23 comments as developers picked over how well the automatic-parallelism claim holds up outside of toy examples. The core sell isn't raw speed on a single core. It's that the same unmodified program scales as more hardware is thrown at it.\n\nParallel programming has stayed hard for decades precisely because reasoning about shared memory, races, and synchronization doesn't map cleanly onto how most people write code. If Bend's interaction-combinator approach genuinely lets ordinary control flow parallelize itself, it lowers the bar for using a GPU for general-purpose computation, not just matrix math. That matters far beyond machine-learning teams already fluent in CUDA; it's aimed at the much larger pool of developers who have never written a GPU kernel.\n\nAutomatic parallelization has been promised before, from Fortran auto-parallelizing compilers in the 1980s to more recent parallel DSLs, and the pattern has usually been the same: beautiful benchmark numbers, messier results on real, branch-heavy programs. Bend's interaction-net approach is a genuinely different bet than those earlier attempts, but the burden of proof still sits with the runtime, not the pitch.","[\"programming-languages\",\"gpu-computing\",\"parallel-computing\",\"open-source\"]","2026-09-17T20:36:13.000Z","2026-09-18T14:03:04.571Z","2026-09-18T14:03:15.849Z","published",null,[],"dev-tools",[26,27,28,29],"programming-languages","gpu-computing","parallel-computing","open-source",[31],{"name":32,"url":33},"Hacker News","https:\u002F\u002Fbend-lang.com\u002F",0,{"sections":36},[37,42,46,51,56,61,65,70,74,79,84,89,94,99],{"name":38,"slug":39,"count":40,"latest_published_at":41},"AI","ai",3910,"2026-09-18T04:00:00.000Z",{"name":43,"slug":44,"count":45,"latest_published_at":41},"Security","security",651,{"name":47,"slug":48,"count":49,"latest_published_at":50},"Policy","policy",338,"2026-09-11T04:00:00.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",154,"2026-09-17T04:00:00.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":60},"Science","science",114,{"name":66,"slug":67,"count":68,"latest_published_at":69},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":71,"slug":24,"count":72,"latest_published_at":73},"Dev Tools",76,"2026-09-18T01:04:54.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":78},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":85,"slug":86,"count":87,"latest_published_at":88},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]