Hugging Face and Earthmover Publish Guide to Running Open AI Weather Models Like AIFS and Aurora
ariG23498 · x · 2026-09-08
Hugging Face and Earthmover released a joint guide to reduce friction in running open-weights AI weather forecasting models. ECMWF's AIFS and Microsoft's Aurora publish weights on HF, and Google DeepMind open-sourced WeatherNext 2 — but open weights don't mean easy to run.
Key points:
- Step-by-step tutorial for running AI weather models locally; inference takes seconds on a laptop or small cluster
- Analysis-ready weather data from the Earthmover Marketplace for initial conditions and validation
- Live demo Space to run different models directly
- Covers practical friction: unified data formats, flash-attention dependencies, rendering outputs
A substantive step toward making open AI weather forecasting accessible to everyone.
Related event: Hugging Face Tutorial Makes Open AI Weather Models Accessible(2 posts)→
More from Infra
- $22.5M of Compute in One Week: The New Price Tag for Math Breakthroughs — vykthur · 2026-09-09
- $1M sounds like a lot — it's just 10.5 minutes of OpenAI's compute spend — wordgrammer · 2026-09-09
- SageMaker Feature Store adds UpdateRecord for atomic feature-level writes, no more read-modify-write — AWS ML Blog · 2026-09-09
- Modular shows unified compute layer: TPU v6e and d-Matrix Corsair brought up in days — carrycooldude · 2026-09-09
- LM Studio ships Bionic 1.1.2 with faster sessions, better screen reader support, Linux builds — mattturck · 2026-09-09
- Navier-Stokes proof burned 300B output tokens, $20-30M at consumer API prices — soumitrashukla9 · 2026-09-09