New algorithms open possibilities for training AI models on analog chips
Research
Peter Hess
A lab for the future of computing
Welcome
In a lab in Yorktown sits the most advanced quantum computer in the world, and a cluster of revolutionary prototype AI chips. It’s where we’re inventing what’s next.
Recent news
Europe’s first IBM Quantum Data Center is now open
NewsScott Crowder and Steffen ThoßFor LLMs, IBM’s NorthPole chip overcomes the tradeoff between speed and efficiency
ResearchPeter HessPonder This:
October ChallengePuzzleHow memory augmentation can improve large language model efficiency and flexibility
ResearchPeter HessIntroducing Prithvi WxC, a new general-purpose AI model for weather and climate
NewsKim Martineau- Read more news on our blog
Tools + Code
InstructLab
A community-based approach to building open-source LLMs, created by engineers from Red Hat and IBM Research.
View project →IBM Granite Open-source Models
A series of enterprise-focused, open-source models for language, code, time series and geospatial.
View project →Project CodeNet
A large-scale dataset with approximately 14 million code samples, each of which is an intended solution to one of 4000 coding problems. Rich annotation enables research in code search, code completion, code-code translation, and myriad other use cases.
View project →GT4SD
An open-source library to accelerate hypothesis generation in the scientific discovery process.
View project →