Silurian is building foundation models for simulating the Earth, starting with weather: silurian.ai

I am happy to announce my #NeurIPS2024 paper: APEBench. Check out the project page: tum-pbs.github.io/apebench-paper/ Code: github.com/tum-pbs/apeben… pip install apebench P.S.: I will be in Vancouver and visit SF for six days before the conference. DM me if you want to chat.

Our Earth API with weather forecasts is out now!

We upgraded our foundation model and our Earth API just in time to make operational predictions of the Gulf Coast winter storm! Read about the upgrade in our blog (link in reply)

It's my longstanding belief that to make better precipitation forecasts, our models need to see the full picture. Excited to get our upgrades to our foundation model GFT and Earth API live for everyone. When we say there's a 70% chance of rain, you can count on it! ⛈️

We learned why everyone hates Google Cloud... No notice at all.

Come join us and help accelerate energy transition.

Silurian is proud to join the TotalEnergies On Acceleration Program! This six-month program provides tailored support, collaboration with TotalEnergies' experts, and access to its infrastructure and customer network.

Silurian is excited to announce we are partnering with the utility company Hydro Québec to support grid reliability through our foundational weather models. Read more about our partnership in our press release. lnkd.in/em4XYEJx

Proud to announce our partnership with HQ to demonstrate how foundation models can help improve grid reliability.

Relying on HRRR? Meet your upgrade. We just launched GFT-US at Silurian AI : faster refresh, longer range, state-of-the-art accuracy, and no waitlists. You can just head to our API to start using it right now!

Just disabled Oracle on Modal. Suddenly jobs don't time out, coffee tastes richer, colors are brighter, and I can hear clouds moving.

Great to see Nature validate our bet on a single foundation model for the entire planet. It's all connected, and our models should be connecting all the data too. Aurora is a milestone, now to turn breakthrough science into daily decisions.

🚀 Introducing PhysiX: One of the first large-scale foundation models for physics simulations! PhysiX is a 4.5B parameter model that unifies a wide range of physical systems, from fluid dynamics to reaction-diffusion, outperforming specialized, state-of-the-art models.

Great to see more effort in the direction of "generalizable" neural simulations!

The Aurora foundation model is now fully open and MIT licensed (including the weights).

An overlooked aspect of forecasting: knowing when a storm will be a non-threat (like Henriette) is just as valuable as knowing when one will become a major hurricane (like Erin). Predicting a dissipation saves resources and prevents false alarms. Predicting a major storm saves

GFT-C has been consistently amazing. x.com/SilurianAI/sta…