[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-framework-speeds-up-soft-robot-simulations-with-gpus":10,"sections":44},{"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":34,"tags":35,"sources":39,"feedback":43,"feedback_at":22,"cost_usd":43,"total_tokens":43},9326,"new-framework-speeds-up-soft-robot-simulations-with-gpus","New Framework Speeds Up Soft Robot Simulations With GPUs","A new open-source tool called SoRoMoX runs soft-robot physics simulations up to 680 times faster by moving the math onto GPUs.","Soft robots just got a much faster brain for simulation and control.\n\nResearchers released SoRoMoX, a Python framework built on JAX that models soft robots using Cosserat-rod theory, the standard math for simulating flexible, continuum-like limbs. Unlike older tools, it runs on GPUs and is fully differentiable, meaning every force, torque, and parameter in a simulation can feed directly into an optimization or learning loop. On a single CPU, SoRoMoX's rollouts run up to 27 times faster than SoRoSim, an established comparison framework; parallelized on a GPU, that speedup climbs to as much as 679.7 times for its general variable-strain model. The same differentiability also improved related tasks: system identification saw 66% lower marker error, computed-torque tracking cut error by roughly 500 times versus basic PD control, and safety-constrained control held contact forces under a 5 N limit instead of 33.5 N without that constraint.\n\nSoft robotics has had workable models for years - the real bottleneck was turning those models into something fast enough to actually train on. SoRoMoX's biggest practical win may be reinforcement learning (RL): training a control policy ran up to 7 times faster than a CPU-based baseline called PyElastica, thanks to massively parallel GPU rollouts. That is the difference between a multi-day training run and an afternoon one.\n\nRigid robots got GPU-accelerated, differentiable physics years ago; soft robots, squishier and harder to model, are only now catching up.","[\"robotics\",\"ai\",\"open-source\",\"reinforcement-learning\"]","2026-10-01T04:00:00.000Z","2026-10-02T09:36:58.573Z","2026-10-02T09:37:00.320Z","published",null,[24,30],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"Remove or qualify the 'open-source' and 'released' framing in the opening — the source abstract never states that SoRoMoX's code has been publicly released or open-sourced, so this is an unsupported claim.","resolved",{"id":31,"reviewer":26,"round":32,"reason":33,"status":29},"editor-r2",2,"The dek uses the acronym 'RL' without ever spelling out 'reinforcement learning (RL)' anywhere in the piece — define it on first use (ideally in the body before\u002Fat the dek's claim) per our acronym-definition rule.","ai",[36,34,37,38],"robotics","open-source","reinforcement-learning",[40],{"name":41,"url":42},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2608.06650",0,{"sections":45},[46,50,55,60,65,70,74,79,84,88,92,97,102,107],{"name":47,"slug":34,"count":48,"latest_published_at":49},"AI",5694,"2026-10-01T12:05:27.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Security","security",821,"2026-10-01T14:00:00.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Policy","policy",431,"2026-10-01T11:08:42.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Deals","deals",311,"2026-10-01T14:19:00.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Hardware","hardware",197,"2026-10-01T11:37:06.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":18},"Science","science",165,{"name":75,"slug":76,"count":77,"latest_published_at":78},"Consumer Tech","consumer-tech",150,"2026-10-01T11:59:27.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Dev Tools","dev-tools",93,"2026-10-01T02:30:48.000Z",{"name":85,"slug":86,"count":82,"latest_published_at":87},"Software","software","2026-09-30T21:41:11.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":54},"Startups","startups",86,{"name":93,"slug":94,"count":95,"latest_published_at":96},"Gaming","gaming",51,"2026-09-30T16:24:30.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"General","general",50,"2026-09-30T21:37:54.000Z",{"name":103,"slug":104,"count":105,"latest_published_at":106},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":108,"slug":109,"count":110,"latest_published_at":111},"How-To","how-to",7,"2026-10-01T09:00:00.000Z"]