Twitxy Saved

UniPat AI

This account is verified because it's an official organization on X. Learn more

@UniPat_AI

Our mission is to accelerate AI's evolution towards real-world automation. Join us unipat.ai.
unipat.ai Joined January 2026
0 Following 877 Followers 54 Posts

Latest posts 5

UniPat AI @UniPat_AI
Proud to collaborate with @DeepPrinciple on the data for Hydrogen-1.0.

Hope to see open-weight models take scientific discovery further.
Chenru Duan Verified account @chenru_duan 6d
We’re introducing Hydrogen-1.0, our first open-weight model for scientific research. We post-trained GLM-5.3 (@Zai_org) on our internal scientific data, lab records, and discovery trajectories.

It scores 38.3% on FrontierScience-Research vs GPT-6-Astra’s 33.3% and scores 93.0% on GPQA-Diamonds.

We believe science should be open, collaborative, and democratized, rather than proprietary. Only open weights let researchers inspect, adapt and build on the unprecedented speed of AI growth. There are many recent models named by chemical elements, but we believe Hydrogen-1.0 is the first aiming towards and open scientific superintelligence for discovery. 🧵
View quoted post on X
UniPat AI @UniPat_AI
Excited to have worked with @ElorianAI on ScienceVision. Huge thanks to @Andrew and everyone who contributed to making this work happen!

It all starts with a question: Can AI observe and think like a scientist?

With ScienceVision, we take this question into the world of scientific 3D objects, exploring whether AI can gather evidence, make sense of what it sees, and reason about it.

More to explore and discover in 3D!
Andrew M. Dai Verified account @AndrewDai 1d
Tired of seeing benchmarks saying we are at AGI but models still can't understand a 3D protein view or a floor plan? When a visual model gets something wrong, is it failing to see or failing to reason about what it sees?

We've built two benchmarks that answers these questions. Explore DecoyCount and ScienceVision here: github.com/elorian-ai/dec… and huggingface.co/datasets/Unipa…

And thank you @mercor and @UniPat_AI for your partnership in building these!
View quoted post on X
1 replies 0 reposts 7 likes 1.1K views View on X
UniPat AI @UniPat_AI
Each discipline contributes 50 questions: biology, chemistry, medical imaging, mechanical CAD and architecture.

Tasks include classifying atomic hybridization, locating a lesion in a CT volume and assessing a part’s symmetry.

Models must use domain knowledge to interpret geometric or topological evidence.
UniPat AI @UniPat_AI
ScienceVision combines scientific 3D questions, inspection tools and expert reference answers across disciplines.
All 250 questions have reference inspection records. Harbor-format oracle solutions support task validation and reproducibility.
Watch how experts inspect 3D objects and gather evidence in the reference inspections below.
0:21