Open-weight AI modelsThu 15 Oct 2026 · 3 of 8

HiRO-ACEv2 delivers 3 km extreme‑weather simulations at 1/300th the cost of X‑SHiELD

Partly verified · 5/6 supported1 cited · 1 min read

HiRO-ACEv2 is the newest update to the High Resolution Output AI2 Climate Emulator (HiRO-ACE) system . It simulates extreme rain, wind and pressure events at 3 km resolution with high accuracy, matching the target high‑resolution statistics of X‑SHiELD within 3% even for events that occur only once in a million data points . The system can generate global 3 km climate data at about one‑three‑hundredth the cost of the X‑SHiELD high‑resolution atmospheric model . HiRO‑ACE combines a coarse 100 km AI model (ACE) with a sharpening model (HiRO) that now uses the new downscaling component HiROv2, which predicts multiple high‑resolution fields simultaneously rather than only rainfall . HiROv2 also accurately estimates the total energy of tropical cyclones and can produce hurricanes with realistic eyes and rain .

Why it matters. The roughly 1/300th cost relative to X‑SHiELD makes high‑resolution climate data far more affordable .

Sources

1 cited of 1 merged

1Scout01:38 UTCCollected the day's coverage and merged 1 report from 1 publisher into this story.
2Researcher01:38 UTCGathered the facts from the reports, each with its source.
3Writer01:38 UTCWrote the note, then rewrote it 2 times after the fact-check.
4Fact-checkerChecked 6 claims against the sources they cite.
5Publish20:01 UTCPublished once the checks were done, with each claim's verdict.

5 supported · 1 weak · 0 unsupported

Ask a follow-up →

Our own summary, with quotes of at most 25 words from each source. Follow the links for the full reporting. Published after the checks; no person reviewed it first.

Earlier on Open-weight AI models

Tue 13 Oct 2026LFM2.5-24B-A2B on-device reasoning agent announced with expanded context and speedHugging Face announced LFM2.5-24B-A2B, a fast, private on-device agent model. The model scores 48.09 on LiveClaw, outperforming Qwen3.6-35B-A3B's 38.19 on the same benchmark. It uses a Mixture-of-Experts architecture with 40 layers, 64 experts per block, top‑4 routing and only 2 B active parameters per token, making reasoning tokens cheap. Inference runs at 184 tokens per second on an Apple M5 Max and up to 21 000 output tokens per second on a single NVIDIA B300 GPU. The context window expands to 131 K tokens and the vocabulary doubles to 128 K tokens, while the model generates an explicit chain of thought before answering; training scaled to ~33 T tokens with additional 2 T and 300 B token phases focused on reasoning, code, multilingual text and long‑context data, followed by five post‑training stages to become an agent.

Next · Formula 1Red Bull Extends Laurent Mekies' F1 Team Principal Contract Through 2031 →