INVA-AI-HERALD
๐๐ก๐ฉ๐-๐๐-๐๐๐ฅ๐๐๐ ๐ฐ ๐๐ ๐ก๐๐ช๐ฆ 1. ๐งช ๐๐ฒ๐ฒ๐ฝ ๐ฃ๐ฟ๐ถ๐ป๐ฐ๐ถ๐ฝ๐น๐ฒ ๐ฅ๐ฒ๐น๐ฒ๐ฎ๐๐ฒ๐ ๐๐๐ฑ๐ฟ๐ผ๐ด๐ฒ๐ป-๐ญ.๐ฌ, ๐ฎ๐ป ๐ข๐ฝ๐ฒ๐ป-๐ช๐ฒ๐ถ๐ด๐ต๐ ๐ฆ๐ฐ๐ถ๐ฒ๐ป๐๐ถ๐ณ๐ถ๐ฐ ๐๐ ๐ ๐ผ๐ฑ๐ฒ๐น ๐ฐ ๐ช๐๐๐ง ๐๐๐ฃ๐ฃ๐๐ก๐๐ !! Deep Principle introduced Hydrogen-1.0, an open-weight scientific model built by post-training Z.ai’s GLM-5.3 on chemistry, materials-science and life-science data, alongside laboratory records and agent trajectories from its own research work. The model is designed to help researchers move beyond isolated question-answering and reason across experimental workflows, including candidate selection, synthesis planning, evaluation and revised hypotheses. Its weights are available through Hugging Face and ModelScope, with API access and integration into the company’s Mira AI Scientist. The full training datasets and complete training recipe have not been released. ๐ง ๐ช๐๐๐ง ๐๐๐ก ๐๐ง ๐๐ข? Researchers can use it to explore chemistry and materials problems, organise expe...