AI & ML interests
agent model, continual learning
Recent Activity
View all activity
Papers
On the Scaling of PEFT: Towards Million Personal Models of Trillion Parameters
Macaron-A2UI: A Model for Generative UI in Personal Agents
Organization Card
Mind Lab is a frontier lab made up of a dedicated team of about 30 researchers.
We build large-scale agent models, along with their training infrastructure and inference systems, and AI agents designed and deployed to interact with users and handle complex, real-world tasks.
We also study continuous learning, exploring how AI learns from real experiences and how diverse populations of models can collaborate to improve overall system intelligence.
Model for Generative UI in Personal Agent
-
mindlab-research/Macaron-A2UI-Tall
Text Generation • Updated • 59 • 4 -
mindlab-research/Macaron-A2UI-Grande
Text Generation • Updated • 46 • 4 -
mindlab-research/Macaron-A2UI-Venti
Text Generation • Updated • 23 • 2 -
Macaron-A2UI: A Model for Generative UI in Personal Agents
Paper • 2605.24830 • Published • 81
Model for Generative UI in Personal Agent
-
mindlab-research/Macaron-A2UI-Tall
Text Generation • Updated • 59 • 4 -
mindlab-research/Macaron-A2UI-Grande
Text Generation • Updated • 46 • 4 -
mindlab-research/Macaron-A2UI-Venti
Text Generation • Updated • 23 • 2 -
Macaron-A2UI: A Model for Generative UI in Personal Agents
Paper • 2605.24830 • Published • 81
models 5
mindlab-research/Macaron-V1-Preview-749B
Text Generation • 754B • Updated • 1k • 46
mindlab-research/Macaron-V1-Preview-744B-Merged
Text Generation • 754B • Updated • 1 • 4
mindlab-research/Macaron-A2UI-Venti
Text Generation • Updated • 23 • 2
mindlab-research/Macaron-A2UI-Grande
Text Generation • Updated • 46 • 4
mindlab-research/Macaron-A2UI-Tall
Text Generation • Updated • 59 • 4