--- title: HiRO-ACEv2: Simulating extreme weather at 3 km... description: Ai2 released HiRO-ACEv2, an updated version of its High Resolution Output Ai2 Climate Emulator that generates global 3 km resolution weather simulations. The... canonical: https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost | daily.dev og:description: Ai2 released HiRO-ACEv2, an updated version of its High Resolution Output Ai2 Climate Emulator that generates global 3 km resolution weather simulations. The... og:url: https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj og:image: https://api.daily.dev/og/posts/CyobP1Ewj.png og:image:alt: HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost 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. # HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost **[Hugging Face](https://daily.dev/sources/huggingface)** · 4 min read · 0 upvotes · 0 comments ## Summary Ai2 released HiRO-ACEv2, an updated version of its High Resolution Output Ai2 Climate Emulator that generates global 3 km resolution weather simulations. The new HiROv2 downscaling model predicts multiple high-resolution fields simultaneously (rain, wind, pressure) rather than just rainfall, and better reproduces rare extreme events, matching X-SHiELD's high-resolution statistics within 3% even for one-in-a-million occurrences. The combined system produces global 3 km climate data at roughly 1/300th the cost of running X-SHiELD directly, with regional runs saving even more. Limitations remain: the system only works with the ACE variant tuned to X-SHiELD and can't yet simulate other climates like a warmer future or historical datasets such as ERA5. Future plans include training on a 4°C-warmer world dataset and adding temperature, snow, and ice outputs. Models, a paper, tutorial notebooks, and a web demo comparing HiROv1, HiROv2, and X-SHiELD outputs are available. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is HiRO-ACEv2 and how does it improve on the previous version? HiRO-ACEv2 is an updated AI climate emulator from Ai2 that downscales coarse 100 km ACE simulations into 3 km resolution weather data. Unlike the prior HiROv1, which only predicted rainfall, HiROv2 predicts rain, wind, and pressure fields simultaneously and better reproduces extremely rare events, matching X-SHiELD's high-resolution statistics within 3% even for one-in-a-million occurrences. _Teams evaluating AI climate emulators can track releases like HiRO-ACEv2 on daily.dev._ ### How much cheaper is HiRO-ACEv2 than running X-SHiELD for generating high-resolution climate data? HiRO-ACEv2 generates global 3 km climate data at roughly 1/300th the cost of running X-SHiELD, a high-resolution atmospheric model that explicitly simulates storms. For single-region simulations, the cost savings can exceed a hundred times more than the global figure, depending on the area size, and a faster streamlined version of HiROv2 is in development. _Researchers weighing simulation cost tradeoffs can follow cost-efficiency benchmarks like this on daily.dev._ ### What are the current limitations of the HiRO-ACEv2 climate emulator? HiRO-ACEv2 only works with the version of ACE tuned to match X-SHiELD, so it cannot yet produce high-resolution data for other climates, such as a warmer future scenario or historical datasets like ERA5. Planned future work includes training on new X-SHiELD datasets simulating a 4°C warmer world and adding outputs like near-surface temperature, snow, and ice. _Developers building on emerging climate models can keep up with evolving capabilities via daily.dev._ ## Similar posts on daily.dev - [Introducing WeatherNext 3, our most advanced and accurate global weather AI model](https://daily.dev/posts/introducing-weathernext-3-our-most-advanced-and-accurate-global-weather-ai-model-0gnv0uaqj) · DeepMind · 0 upvotes · 0 comments - [Gen AI Super-resolution Accelerates Weather Prediction with Scalable, Low-Compute Models](https://daily.dev/posts/gen-ai-super-resolution-accelerates-weather-prediction-with-scalable-low-compute-models-ra5xb4rsb) · NVIDIA Developer · 0 upvotes · 0 comments - [Researchers improve weather forecasts looking two to six weeks out](https://daily.dev/posts/researchers-improve-weather-forecasts-looking-two-to-six-weeks-out-bql5ebjsp) · MIT News · 1 upvotes · 0 comments - [This AI weather startup is out-forecasting government agencies](https://daily.dev/posts/this-ai-weather-startup-is-out-forecasting-government-agencies-eiehjy3rs) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#machine-learning](https://daily.dev/tags/machine-learning) [View this post on daily.dev](https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj) ```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":"HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost","url":"https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj"},"datePublished":"2026-10-14T16:14:53.290Z","dateModified":"2026-10-14T16:15:56.750Z","description":"Ai2 released HiRO-ACEv2, an updated version of its High Resolution Output Ai2 Climate Emulator that generates global 3 km resolution weather simulations. The...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a64d2cb3f578699fc58d38c1f23a3742?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a64d2cb3f578699fc58d38c1f23a3742?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hugging Face","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":"Hugging Face","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/f1f55c67d81a4330acf5b90b26b0c8e1","url":"https://daily.dev/sources/huggingface"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"machine-learning","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hugging Face","item":"https://daily.dev/sources/huggingface"},{"@type":"ListItem","position":3,"name":"HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/hiro-acev2-simulating-extreme-weather-at-3-km-resolution-for-a-fraction-of-the-cost-cyobp1ewj#faq","mainEntity":[{"@type":"Question","name":"What is HiRO-ACEv2 and how does it improve on the previous version?","acceptedAnswer":{"@type":"Answer","text":"HiRO-ACEv2 is an updated AI climate emulator from Ai2 that downscales coarse 100 km ACE simulations into 3 km resolution weather data. Unlike the prior HiROv1, which only predicted rainfall, HiROv2 predicts rain, wind, and pressure fields simultaneously and better reproduces extremely rare events, matching X-SHiELD's high-resolution statistics within 3% even for one-in-a-million occurrences. Teams evaluating AI climate emulators can track releases like HiRO-ACEv2 on daily.dev."}},{"@type":"Question","name":"How much cheaper is HiRO-ACEv2 than running X-SHiELD for generating high-resolution climate data?","acceptedAnswer":{"@type":"Answer","text":"HiRO-ACEv2 generates global 3 km climate data at roughly 1/300th the cost of running X-SHiELD, a high-resolution atmospheric model that explicitly simulates storms. For single-region simulations, the cost savings can exceed a hundred times more than the global figure, depending on the area size, and a faster streamlined version of HiROv2 is in development. Researchers weighing simulation cost tradeoffs can follow cost-efficiency benchmarks like this on daily.dev."}},{"@type":"Question","name":"What are the current limitations of the HiRO-ACEv2 climate emulator?","acceptedAnswer":{"@type":"Answer","text":"HiRO-ACEv2 only works with the version of ACE tuned to match X-SHiELD, so it cannot yet produce high-resolution data for other climates, such as a warmer future scenario or historical datasets like ERA5. Planned future work includes training on new X-SHiELD datasets simulating a 4°C warmer world and adding outputs like near-surface temperature, snow, and ice. Developers building on emerging climate models can keep up with evolving capabilities via daily.dev."}}]} ```