---
title: "DeepMimic's Creator Ships MimicKit to Fix Motion AI's Reproducibility Crisis"
subtitle: "Xue Bin Peng released a clean reinforcement learning codebase bundling DeepMimic, AMP, ASE and a new diffusion-based motion prior called SMP."
url: "https://alphasignal.ai/news/deepmimic-s-creator-ships-mimickit-to-fix-motion-ai-s-reproducibility-crisis"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-05-05T15:40:49.000Z"
updated: "2026-08-22T00:38:03.338Z"
type: "Repo"
topics:
  - "Robotics"
  - "Open Source"
subtopics:
  - "Vla Models"
  - "Manipulation"
  - "Humanoid"
upvotes: 16
source_url: "https://github.com/xbpeng/MimicKit"
comments: 0
truncated: true
---

# DeepMimic's Creator Ships MimicKit to Fix Motion AI's Reproducibility Crisis

> Xue Bin Peng released a clean reinforcement learning codebase bundling DeepMimic, AMP, ASE and a new diffusion-based motion prior called SMP.

Motion imitation research has a reproducibility problem. Every paper reimplements DeepMimic or AMP slightly differently, small numerical details silently ruin training, and the reference codebases tend to bit-rot around specific simulator versions. [MimicKit](https://github.com/xbpeng/MimicKit), from Xue Bin Peng (the author behind DeepMimic and AMP), ships a single lightweight framework with correct implementations of the whole family of methods, plus a new one built on diffusion models.

[The repo](https://github.com/xbpeng/MimicKit) describes itself as a suite of motion imitation methods for training motion controllers, designed to be clean and lightweight, with minimal dependencies. It is Apache-2.0 licensed, Python-only, and sitting at around 2.1k stars. The associated [Starter Guide paper](https://arxiv.org/abs/2510.13794) walks through the algorithms in more depth.

![Simulated Unitree Go2 trotting alongside a reference motion clip](https://github.com/xbpeng/MimicKit/raw/main/images/teaser_go2.gif) 

## Every mocap-to-controller method under one roof

Rather than scattering across seven repos, MimicKit exposes a consistent CLI and config system for the classic and current motion imitation algorithms:

* **DeepMimic**: the original tracking-reward approach that made humanoids do spinkicks and backflips.
* **AMP** (Adversarial Motion Priors) and **ASE** (Adversarial Skill Embeddings): GAN-style discriminators trained against mocap so a policy learns stylistic naturalness without exact frame tracking.
* **AWR** (Advantage-Weighted Regression),

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [xbpeng/MimicKit GitHub repository](https://github.com/xbpeng/MimicKit)
- [SMP: Reusable Score-Matching Motion Priors for Physics-Based Character Control](https://arxiv.org/abs/2512.03028)
- [SMP README in MimicKit](https://github.com/xbpeng/MimicKit/blob/main/docs/README_SMP.md)
- [MimicKit Starter Guide](https://arxiv.org/abs/2510.13794)
- [SMP project page](https://yxmu.foo/smp-page/)