Watch how Google DeepMind chemists used Antigravity to extract compound data from patents and papers, analyze what drives potency, and propose new candidate molecules with synthesis routes in just an hour. The agent also launches an interactive viewer so chemists can explore the data in a collaborative way: browsing... retrieved molecules, inspecting their properties, and plotting how molecular structure shapes potency.show more

For the full deep-dive:

@GoogleDeepMind Wait it can dig through patents and research papers and create interactive viewers? My chemistry teacher is crying in the corner lol

@GoogleDeepMind the interactive viewer is the part i'd steal

@GoogleDeepMind we new carbon and hydrogen models and generous usage limits and fast models because argon very slow in antigravity. right?

@GoogleDeepMind Both looks like AI avatar or they are real ?

@GoogleDeepMind Mind blowing from patents to new candidate molecules with synthesis routes in just an hour. Antigravity is supercharging real science!

@GoogleDeepMind Impressive how AI agents are speeding up complex research and molecular discovery.

@GoogleDeepMind people spend weeks digging through old chemistry patents while an agent just reads them all and builds a custom viewer before lunch

@GoogleDeepMind Chemists just got the molecule and the recipe in one go.

@GoogleDeepMind Antigravity chemistry workflow looks incredible, turning patents into candidates hours

@GoogleDeepMind Antigravity turning patent data into new molecule ideas in an hour sounds truly impressive

@GoogleDeepMind From patents to new molecules with synthesis routes in an hour Antigravity is turning chemistry research into real time exploration

@GoogleDeepMind From “can you summarize this PDF?” to “here are some candidate molecules.” The follow-up questions are getting ambitious

@GoogleDeepMind Plotting structure against potency, because apparently chemistry needed a dashboard now.

@GoogleDeepMind umgehen die damit patente oder liest sich dass nur so ;)

@GoogleDeepMind the hour holds until a table image hides the units, and a potency number read wrong looks the same as one read right.

@GoogleDeepMind 一小时从专利论文里挖数据、分析活性、给出候选分子和合成路线，这个速度对化学家来说确实是质变，交互查看器也很实用。

@GoogleDeepMind This is the future of AI-powered chemistry! 🧪 Turning hours of research into actionable molecular insights in just one hour is incredible. 🚀

@GoogleDeepMind I wouldn't trust anything coming from Gemini models.

@GoogleDeepMind This very clearly explains why these chemists named their models that way

@GoogleDeepMind Pairing potency analysis with proposed synthesis routes gives chemists a concrete starting point for choosing what to make and test next.

@GoogleDeepMind Advanced models are in high demand right now, so a different model was used for this response. This didn’t count toward your limit.