# Ai2's HiRO-ACEv2 Generates 3-Kilometer Global Extreme-Weather Data at About 1/300th the Cost of the Physics Model It Emulates

The upgraded downscaler predicts rain, wind and pressure together and, according to Ai2, matches the target model's statistics within 3% even for events that occur once in a million data points.

- Published: 2026-10-21T06:08:03.066Z
- Canonical: https://polylog.news/ai/2026-10-21/ai2-s-hiro-acev2-generates-3-kilometer-global-extreme-weathe
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news)

The Allen Institute for AI (Ai2) released HiRO-ACEv2, a new version of its climate emulator. The Russian-language channel AI ML Big Data reported that it reproduces heavy rain, hurricane-force wind and pressure swings at 3-kilometer detail.…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-10-21/ai2-s-hiro-acev2-generates-3-kilometer-global-extreme-weathe (subscription information: https://polylog.news/pricing).