# lzyang2000/morphmimic

> Research project: trains motion-tracking policies for multiple humanoids (Unitree G1, AgiBot X2 Ultra, Booster T2) via a differentiable world model, sharing the world model and/or policy across robots with a morphology vector.

- Magnitude: 3.7 out of 10 — Early signal
- Stars: 134 total · +14 stars measured 2026-10-27, 01:17–05:33 UTC, ≈ 44 by evening
- Star trust: star growth looks organic
- Category: Hardware, IoT & robotics · Language: Python · License: Apache-2.0 · Created: 2026-10-23 · Last push: 2026-10-26
- GitHub: https://github.com/lzyang2000/morphmimic · Page: https://gitnova.dev/en/r/lzyang2000/morphmimic

## Useful for

- Train a motion-tracking policy for G1, X2 Ultra and T2 on the BONES-SEED dataset
- Export a checkpoint to ONNX and run sim2sim over ZMQ with MuJoCo
- Teleoperate a humanoid via a PICO VR headset with GMR retargeting

## Why it’s here

- Star-counter measurements on 2026-10-27 (UTC), 01:17–05:33: 120 → 134 stars (+14). This is the change over that interval.
- Estimated end-of-day forecast: about +44 stars, using observed gains and the previous day.
- The repository is 4 days old and already has 134 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 12
- Issues and pull requests: 0
- Watchers: 0
- Average over the last week: 36 per day
- Usual pace: too little history (under two weeks)
- Stars in the last hour (measured): 1

## Stars per day, last 5 days (oldest → newest, today is partial)

2026-10-23 … 2026-10-27: 17, 39, 44, 34, 14

## Similar by description

1. **CuboiLeo/SuperMimic-Reproduced** — 2.7 · Cooling · Science & research · Python · +4 stars measured 2026-10-27, 00:37–05:35 UTC, ≈ 20 by evening
   From-scratch reproduction of the SuperMimic paper (arXiv:2610.28480): a differentiable world model plus short-horizon BPTT actor-critic for Unitree G1 whole-body motion tracking, built on mjlab/MuJoCo Warp, with data pipeline, training,…
   Full card: https://gitnova.dev/en/r/CuboiLeo/SuperMimic-Reproduced.md
2. **Motphys/Uni-SuperMimic** — 2.3 · Early signal · Hardware, IoT & robotics · Python · +5 stars measured 2026-10-27, 00:39–05:36 UTC, ≈ 19 by evening
   An independent reproduction of SuperMimic on UniLab for whole-body motion tracking of the Unitree G1 humanoid on LAFAN1 data, with mjwarp, MuJoCo, and Motrix backends. Includes a differentiable world model, a goal-conditioned policy, and…
   Full card: https://gitnova.dev/en/r/Motphys/Uni-SuperMimic.md
3. **pollen-robotics/microduck_rl** — 2.8 · Breakout · Hardware, IoT & robotics · Python · +9 stars measured 2026-10-27, 00:37–05:35 UTC, ≈ 38 by evening
   Reinforcement learning environments for the Microduck bipedal robot built on mjlab (MuJoCo Warp) with PPO: train walking and trick policies, export to ONNX, and deploy to the real robot.
   Full card: https://gitnova.dev/en/r/pollen-robotics/microduck_rl.md
4. **DiaoOrg/awesome-supermimic** — 5.5 · Early signal · Learning & lists · +29 stars measured 2026-10-27, 00:34–05:32 UTC, ≈ 121 by evening
   A curated list of implementations, extensions and demos of SuperMimic, a method for humanoid whole-body control via a differentiable world model. It serves as an entry point for reproducing and adapting the method.
   Full card: https://gitnova.dev/en/r/DiaoOrg/awesome-supermimic.md
5. **zita-ch/supermimic_vibed** — 4.0 · Cooling · Hardware, IoT & robotics · Python · +14 stars measured 2026-10-27, 00:35–05:33 UTC, ≈ 58 by evening
   Unofficial reproduction of SuperMimic — training a single humanoid G1 whole-body control policy to track 40 LAFAN1 clips on Isaac Lab 3 with MuJoCo-Warp physics, reproducing the paper's experiment.
   Full card: https://gitnova.dev/en/r/zita-ch/supermimic_vibed.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-27 05:43 UTC, updated every 30 minutes.