The U.S. Department of Energy has launched the Genesis Open Models Initiative and, in partnership with Arcee, unveiled its first open-weight model for scientific research named Genesis-Science-1.
No further details regarding specific capabilities or benchmarks are provided in the source text.
Google DeepMind and Google Research have open sourced the WeatherNext 2 and WeatherNext Cyclones AI models, which achieved state-of-the-art accuracy in predicting tropical cyclone tracks, intensity, and wind structure. Published in Nature, the work demonstrates that these models provide forecasters with an extra day of predictive accuracy compared to prior systems.
Moonshot AI has introduced Kimi K3, an open-source Mixture-of-Experts model featuring 2.8 trillion total parameters and 104 billion activated parameters per token. The model supports native vision capabilities and a 1-million-token context window, leveraging Kimi Delta Attention and Stable LatentMoE to achieve approximately 2.5x scaling efficiency over its predecessor, Kimi K2.
Moonshot AI introduces Kimi K3, an open-source Mixture-of-Experts model featuring 2.8 trillion total parameters and 104 billion activated parameters per token. The architecture utilizes native vision capabilities and supports a 1-million-token context window, enabled by Kimi Delta Attention and Stable LatentMoE.
Arcee, a developer of open-weight models in the US, has partnered with the Department of Energy to create Genesis-Science-1. This collaboration results in an open model designed specifically for scientific research.
The Department of Energy (DOE) and Arcee AI have announced the development of Genesis-Science-1 (GS1), an open-weight language model designed specifically for scientific research. This initiative is part of the Genesis Mission, aiming to bring advanced AI capabilities to a wide range of scientific fields through a joint effort between the government agency and the private company.
We use cookies to measure traffic and improve the site. You can accept or decline analytics cookies. Privacy policy