# Stable Baselines Jax (SB3 + JAX = SBX)
See https://github.com/araffin/sbx
Proof of concept version of [Stable-Baselines3](https://github.com/DLR-RM/stable-baselines3) in Jax.
Implemented algorithms:
- [Soft Actor-Critic (SAC)](https://arxiv.org/abs/1801.01290) and [SAC-N](https://arxiv.org/abs/2110.01548)
- [Truncated Quantile Critics (TQC)](https://arxiv.org/abs/2005.04269)
- [Dropout Q-Functions for Doubly Efficient Reinforcement Learning (DroQ)](https://openreview.net/forum?id=xCVJMsPv3RT)
- [Proximal Policy Optimization (PPO)](https://arxiv.org/abs/1707.06347)
- [Deep Q Network (DQN)](https://arxiv.org/abs/1312.5602)
- [Twin Delayed DDPG (TD3)](https://arxiv.org/abs/1802.09477)
- [Deep Deterministic Policy Gradient (DDPG)](https://arxiv.org/abs/1509.02971)
- [Batch Normalization in Deep Reinforcement Learning (CrossQ)](https://openreview.net/forum?id=PczQtTsTIX)
- [Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning (SimBa)](https://openreview.net/forum?id=jXLiDKsuDo)
## Example
```python
from sbx import DDPG, DQN, PPO, SAC, TD3, TQC, CrossQ
model = TQC("MlpPolicy", "Pendulum-v1", verbose=1)
model.learn(total_timesteps=10_000, progress_bar=True)
Raw data
{
"_id": null,
"home_page": "https://github.com/araffin/sbx",
"name": "sbx-rl",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "reinforcement-learning-algorithms reinforcement-learning machine-learning gym gymnasium jax openai stable baselines toolbox python data-science",
"author": "Antonin Raffin",
"author_email": "antonin.raffin@dlr.de",
"download_url": "https://files.pythonhosted.org/packages/00/85/a65bd86e95e0606fe00c971db7edf826969f977ef0c6ff07e5b85ffac791/sbx_rl-0.22.0.tar.gz",
"platform": null,
"description": "\n\n# Stable Baselines Jax (SB3 + JAX = SBX)\n\nSee https://github.com/araffin/sbx\n\nProof of concept version of [Stable-Baselines3](https://github.com/DLR-RM/stable-baselines3) in Jax.\n\nImplemented algorithms:\n- [Soft Actor-Critic (SAC)](https://arxiv.org/abs/1801.01290) and [SAC-N](https://arxiv.org/abs/2110.01548)\n- [Truncated Quantile Critics (TQC)](https://arxiv.org/abs/2005.04269)\n- [Dropout Q-Functions for Doubly Efficient Reinforcement Learning (DroQ)](https://openreview.net/forum?id=xCVJMsPv3RT)\n- [Proximal Policy Optimization (PPO)](https://arxiv.org/abs/1707.06347)\n- [Deep Q Network (DQN)](https://arxiv.org/abs/1312.5602)\n- [Twin Delayed DDPG (TD3)](https://arxiv.org/abs/1802.09477)\n- [Deep Deterministic Policy Gradient (DDPG)](https://arxiv.org/abs/1509.02971)\n- [Batch Normalization in Deep Reinforcement Learning (CrossQ)](https://openreview.net/forum?id=PczQtTsTIX)\n- [Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning (SimBa)](https://openreview.net/forum?id=jXLiDKsuDo)\n\n## Example\n\n```python\nfrom sbx import DDPG, DQN, PPO, SAC, TD3, TQC, CrossQ\n\nmodel = TQC(\"MlpPolicy\", \"Pendulum-v1\", verbose=1)\nmodel.learn(total_timesteps=10_000, progress_bar=True)\n\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Jax version of Stable Baselines, implementations of reinforcement learning algorithms.",
"version": "0.22.0",
"project_urls": {
"Homepage": "https://github.com/araffin/sbx"
},
"split_keywords": [
"reinforcement-learning-algorithms",
"reinforcement-learning",
"machine-learning",
"gym",
"gymnasium",
"jax",
"openai",
"stable",
"baselines",
"toolbox",
"python",
"data-science"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "5808ba7374d3ba9e3e8aabeb52e20fbcd69c174ad3b6ec4960084f69c1e0b347",
"md5": "bd974ff099e2b99e03ec9ad735413b94",
"sha256": "2d45fed797da97ceaca637f13a5753c1293d38b4fb703e06ba9492287b723ca0"
},
"downloads": -1,
"filename": "sbx_rl-0.22.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "bd974ff099e2b99e03ec9ad735413b94",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 61600,
"upload_time": "2025-07-25T09:34:50",
"upload_time_iso_8601": "2025-07-25T09:34:50.888748Z",
"url": "https://files.pythonhosted.org/packages/58/08/ba7374d3ba9e3e8aabeb52e20fbcd69c174ad3b6ec4960084f69c1e0b347/sbx_rl-0.22.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "0085a65bd86e95e0606fe00c971db7edf826969f977ef0c6ff07e5b85ffac791",
"md5": "0529b6b1803237cce3c825ca7fe6a3e9",
"sha256": "0bbec4a84e23aa616ee66d1bc70a671e194a92f0d23ac9ba74967a7f2867052c"
},
"downloads": -1,
"filename": "sbx_rl-0.22.0.tar.gz",
"has_sig": false,
"md5_digest": "0529b6b1803237cce3c825ca7fe6a3e9",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 50737,
"upload_time": "2025-07-25T09:34:53",
"upload_time_iso_8601": "2025-07-25T09:34:53.117154Z",
"url": "https://files.pythonhosted.org/packages/00/85/a65bd86e95e0606fe00c971db7edf826969f977ef0c6ff07e5b85ffac791/sbx_rl-0.22.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-25 09:34:53",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "araffin",
"github_project": "sbx",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "sbx-rl"
}