Skip to content

Releases: google-deepmind/mujoco_playground

Release v0.1.0

08 Jan 22:18

Choose a tag to compare

[0.1.0] - 2026-01-08

  • Pass through the MuJoCo Warp
    (MjWarp) implementation to MJX, so that MuJoCo Playground environments can
    train with MuJoCo Warp! You can pass through the implementation via the config
    override
    registry.load('CartpoleBalance', config_overrides={'impl': 'warp'}).
  • Update environments to utilize contact sensors and remove collision.py.
  • Remove mjx_env.init in favor of mjx_env.make_data since make_data
    now requires an MjModel argument rather than an mjx.Model argument.
  • Add device to mjx_env.make_data, fixes #174.
  • Update AutoResetWrapper to allow full resets on done. Fixes #179. Also
    provides a means for doing curriculum learning via
    state.info['AutoResetWrapper_done_count'], see #140.
  • Update dependencies to use mujoco>=3.4 and warp-lang>=1.11.

Release v0.0.5

23 Jun 18:39

Choose a tag to compare

[0.0.5] - 2025-06-23

  • Change light_directional to light_type following MuJoCo API change from version 3.3.2 to 3.3.3. Fixes #142.
  • Fix bug in get_qpos_ids.
  • Implement render in Wrapper.
  • Fix #123.
  • Fix #126.
  • Fix #41.

Release v0.0.4

07 Mar 19:25

Choose a tag to compare

Added

  • Added ALOHA handover task (thanks to @Andrew-Luo1).
  • Added Booster T1 joystick task.
booster.mp4
404879775-e8349789-be62-4847-8cb9-1721295f2746.mp4

Changed

  • Fixed foot friction randomization for G1 tasks.
  • Fix various bugs in train_jax_ppo.py (thanks to @vincentzhang).
  • Fixed a small bug in the privileged state of the go1 joystick task.

Release v0.0.3

19 Jan 02:11

Choose a tag to compare

Bumps minimum Python version to 3.10.

Release v0.0.2

17 Jan 00:04

Choose a tag to compare

Initial release.