IBM and NASA launch Surya Artificial Intelligence model to predict solar weather

IBM and NASA unveiled Surya, an open-source Artificial Intelligence model trained on nine years of high-resolution data to forecast solar flares. The model improves solar flare classification accuracy by 16% and is available on Hugging Face and GitHub.

IBM and NASA have collaborated to build Surya, an open-source Artificial Intelligence model designed to forecast solar flares and other solar weather events. Surya was trained on more than nine years of high-resolution imagery from NASA´s Solar Dynamics Observatory. The dataset used for training contains images roughly 10 times larger than typical AI training inputs, enabling finer analysis of solar features that affect Earth´s technological systems.

Technically, the project required bespoke engineering to process the massive images. IBM developed a customized multi-architecture system to handle the scale and detail of the Solar Dynamics Observatory data. According to the article, Surya achieves a 16% improvement in solar flare classification accuracy over previous models, which translates to earlier and more reliable warnings for satellites, communication networks and power grids. The model´s design and training emphasize both predictive performance and operational lead time for mitigation.

Surya is released in an open-source format on platforms such as Hugging Face and GitHub. The availability of models and datasets is intended to encourage international scientific collaboration, enable researchers and developers to refine the system, and broaden applications beyond heliophysics. Public and expert reactions cited in the article include enthusiasm on social media platforms and commentary that the project could extend to earth observation, planetary science and industry uses such as agritech.

The article also frames Surya as having potential economic, social and political implications. Economically, improved forecasting can reduce damage and outages tied to solar storms. Socially, the open-source release is positioned as a tool for education and broader public engagement. Politically, reliable solar weather forecasts are described as relevant to national security because they can help preserve military communications and GPS during geomagnetic events. The piece concludes that Surya’s open approach aims to accelerate research and resilience as the sun approaches a period of increased activity.

75

Impact Score

Saudi Artificial Intelligence startup launches Arabic LLM

Misraj Artificial Intelligence unveiled Kawn, an Arabic large language model, at AWS re:Invent and launched Workforces, a platform for creating and managing Artificial Intelligence agents for enterprises and public institutions.

Introducing Mistral 3: open artificial intelligence models

Mistral 3 is a family of open, multimodal and multilingual Artificial Intelligence models that includes three Ministral edge models and a sparse Mistral Large 3 trained with 41B active and 675B total parameters, released under the Apache 2.0 license.

NVIDIA and Mistral Artificial Intelligence partner to accelerate new family of open models

NVIDIA and Mistral Artificial Intelligence announced a partnership to optimize the Mistral 3 family of open-source multilingual, multimodal models across NVIDIA supercomputing and edge platforms. The collaboration highlights Mistral Large 3, a mixture-of-experts model designed to improve efficiency and accuracy for enterprise artificial intelligence deployments starting Tuesday, Dec. 2.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.