← all repositories

HJYao00/Mulberry

Mulberry is a multi-modal large language model implementing o1-like reasoning and reflection via Collective Monte Carlo Tree Search.

1.2k stars Python Language ModelsAgents
Mulberry
Not currently ranked — collecting fresh signals.
star history

Mulberry is a research project from Nanyang Technological University, Tsinghua University, and Baidu that develops an o1-like reasoning MLLM using Collective MCTS (Monte Carlo Tree Search). The approach enables multi-step reasoning and self-reflection by exploring reasoning paths via tree search. The project includes models available on HuggingFace along with associated training datasets.

Frequently asked

What is HJYao00/Mulberry?
Mulberry is a multi-modal large language model implementing o1-like reasoning and reflection via Collective Monte Carlo Tree Search.
Is Mulberry open source?
Yes — HJYao00/Mulberry is an open-source project tracked on heatdrop.
What language is Mulberry written in?
HJYao00/Mulberry is primarily written in Python.
How popular is Mulberry?
HJYao00/Mulberry has 1.2k stars on GitHub.
Where can I find Mulberry?
HJYao00/Mulberry is on GitHub at https://github.com/HJYao00/Mulberry.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.