mit-han-lab/pear
Library
Can I trust it?
Worth a closer look before you use it.
Warnings Check
Very new: it was created 9 days ago. Wait a little before installing it.
Stars from real accounts No data
We haven't checked its stars for fakes yet.
Active OK
Last change today.
Licence OK
Apache-2.0: you can use it for free, also in a commercial product. Keep the licence notice.
Quick checks from public data. They help, but they are no guarantee.
What it is
PEAR is a post-training engine for agentic robotics that trains reasoning models to use robots through reinforcement learning in simulation. It uses asynchronous GRPO on veRL, with MuJoCo and Isaac Sim simulators on a Ray cluster, and deploys trained models with a vLLM model…
Who it's for
Researchers and engineers training reasoning models for robot manipulation with reinforcement learning in simulation
How to use it
Install
bash scripts/install.sh libero_runFrom its README. Check it there before running it.
Ask your AI to set it up
Paste this into Claude, ChatGPT or your coding assistant.
Help me install and set up mit-han-lab/pear in my project. Follow its README: https://github.com/mit-han-lab/pear. Explain each step and check it is safe before running anything.
Star curve
On the radar
Its place among the radar's top 100, day by day, with #1 at the top. Gaps are days it was out of the top 100.
Records
- Best place on RepoAlerts: #39, on 16 Nov 2026
- In the top 100 for 4 days since 16 Nov 2026
Alternatives
Similar projects named in its profile. Each link searches the directory.
Get alerts for repos like this
We'll email you when something in AI agents starts rising. Free, once a week, only what matches.
What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy
For maintainers
Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.
[](https://repoalerts.com/repo/mit-han-lab/pear?ref=badge)Similar repos
- NVlabs/ASPIRE
A research framework for robotics in which an LLM-driven coding agent refines sensorimotor skills through iterative debugging and evolutionary search instead of gradient-based training. Evaluated on L
- ArcheOne-AI/RoboScript
Agent framework for robot manipulation that learns a task from one demonstration by rehearsing it in a simulator and storing the learned knowledge in the model's context rather than in weights. Valida
- datawhalechina/agentic-rl-lab
A collection of reproductions and tutorials of reinforcement learning algorithms for LLM agents, including GRPO, GSPO, DAPO, OPD, OPSD, Search-R1, ReTool and ALFWorld. Each method is explained and imp
- apple-aiml-research/ml-halo
Research code for the HALO paper on perceptive humanoid motion learned from simulation. It covers synthetic data generation (Infinigen scenes, Kimodo motion, Isaac and MuJoCo rendering), training of a
- Agentic-Intelligence-Lab/CleanRUA
A collection of single-file Python implementations of robot-use agents, where each file contains the prompt, tools and interaction loop of one agent. Agents include Direct, Code, Delegate and Monitor
- shengshu-ai/Motus2
A research world model for robotic dexterous manipulation that uses one shared-weight model as a policy proposing action chunks, a simulator predicting their visual outcomes, and an evaluator scoring