Metadata-Version: 2.4
Name: rlmesh
Version: 0.1.0rc3
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Programming Language :: Rust
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Dist: gym>=0.22.0 ; extra == 'gym'
Requires-Dist: gymnasium>=0.26.1 ; extra == 'gymnasium'
Requires-Dist: huggingface-hub ; extra == 'hf'
Requires-Dist: jax>=0.4.24 ; extra == 'jax'
Requires-Dist: jaxlib>=0.4.24 ; extra == 'jax'
Requires-Dist: numpy>=1.22 ; extra == 'numpy'
Requires-Dist: torch>=1.11 ; extra == 'torch'
Provides-Extra: gym
Provides-Extra: gymnasium
Provides-Extra: hf
Provides-Extra: jax
Provides-Extra: numpy
Provides-Extra: torch
License-File: LICENSE-MIT
License-File: LICENSE-APACHE
License-File: THIRD_PARTY_NOTICES.md
Summary: Python SDK for model-environment evaluation workflows
Keywords: machine-learning,evaluation,model-evaluation,benchmarking,environments,runtime
Home-Page: https://rlmesh.dev
Author-email: "ArenaX Labs, Inc." <research@competesai.com>
Maintainer-email: Josh Blayone <josh@arenaxlabs.com>
License-Expression: MIT OR Apache-2.0
Requires-Python: >=3.10
Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
Project-URL: Documentation, https://docs.rlmesh.dev
Project-URL: Homepage, https://rlmesh.dev
Project-URL: Issues, https://github.com/ArenaX-Labs/rlmesh/issues
Project-URL: Repository, https://github.com/ArenaX-Labs/rlmesh

# RLMesh

RLMesh is a Python SDK for model-environment evaluation. It serves Gymnasium-style environments, connects clients over local or remote transports, and adapts values for plain Python, NumPy, Torch, and JAX users.

> Pre-1.0 (`0.x`): the stable API may change in a minor release, with a migration note, so pin a minor range for active projects.

## Installation

Install from PyPI:

```bash
pip install rlmesh
```

Install optional adapters as needed:

```bash
pip install "rlmesh[numpy]"
pip install "rlmesh[gymnasium]"
pip install "rlmesh[torch]"
pip install "rlmesh[jax]"
```

## Quickstart

Install RLMesh with Gymnasium support and the NumPy client adapter:

```bash
pip install "rlmesh[gymnasium,numpy]"
```

In one process, serve any Gymnasium-compatible environment:

```python
import gymnasium as gym
from rlmesh import EnvServer

env = gym.make("CartPole-v1")
EnvServer(env, "127.0.0.1:5555").serve()
```

In another process, connect to it as a remote environment:

```python
from rlmesh.numpy import RemoteEnv

env = RemoteEnv("127.0.0.1:5555")
observation, info = env.reset(seed=0)

terminated = truncated = False
while not (terminated or truncated):
    action = env.action_space.sample()
    observation, reward, terminated, truncated, info = env.step(action)

env.close()
```

## Links

- Homepage: https://rlmesh.dev
- Documentation: https://docs.rlmesh.dev
- Repository: https://github.com/ArenaX-Labs/rlmesh
- Examples: https://github.com/ArenaX-Labs/rlmesh/tree/main/examples/python
- Issues: https://github.com/ArenaX-Labs/rlmesh/issues
- Default contact: research@competesai.com

## License

Licensed under either of Apache License, Version 2.0 or the MIT license, at your option. See [LICENSE-APACHE](https://github.com/ArenaX-Labs/rlmesh/blob/main/python/rlmesh/LICENSE-APACHE) and [LICENSE-MIT](https://github.com/ArenaX-Labs/rlmesh/blob/main/python/rlmesh/LICENSE-MIT).

Python wheels also include third-party notices in [THIRD_PARTY_NOTICES.md](https://github.com/ArenaX-Labs/rlmesh/blob/main/python/rlmesh/THIRD_PARTY_NOTICES.md).

