---
title: "GalilAI's stable-worldmodel Gives Yann LeCun-Backed AI Researchers One Unified Training Platform"
subtitle: "A new open-source PyTorch platform unifies data collection, training, and MPC-based evaluation for world models across 30+ standardized environments."
url: "https://alphasignal.ai/news/galilai-s-stable-worldmodel-gives-yann-lecun-backed-ai-researchers-one-unified"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-05-30T16:51:02.000Z"
updated: "2026-08-22T03:31:53.673Z"
type: "Repo"
topics:
  - "Video"
  - "Open Source"
subtopics:
  - "World Models"
  - "Video Generation"
  - "Video Understanding"
upvotes: 3550
source_url: "https://github.com/galilai-group/stable-worldmodel"
comments: 0
truncated: true
---

# GalilAI's stable-worldmodel Gives Yann LeCun-Backed AI Researchers One Unified Training Platform

> A new open-source PyTorch platform unifies data collection, training, and MPC-based evaluation for world models across 30+ standardized environments.

World model research has a reproducibility problem. Every new paper re-implements the same baselines, cobbles together its own evaluation protocol, and reinvents data loading pipelines. [stable-worldmodel](https://github.com/galilai-group/stable-worldmodel) is a new open-source PyTorch platform from the GalilAI group that tries to end the reinvention loop by giving researchers one unified toolkit for the entire loop: collect data, train a model, then evaluate it with model-predictive control.

The project has picked up over 2,000 GitHub stars, and its [accompanying paper](https://arxiv.org/abs/2605.21800v1) lists a heavyweight author roster including Yann LeCun and Randall Balestriero. It ships under MIT license and installs with a single `pip install 'stable-worldmodel[data]'`.

## One API for the whole world model loop

A world model is a learned simulator that predicts how an environment evolves in response to actions, letting agents plan by imagining future outcomes. The library exposes that loop through three primitives: a `World` for collecting data, a dataset loader, and a `WorldModelPolicy` wrapping a planning solver. Here is the entire evaluation flow:

```
import stable_worldmodel as swm
from stable_worldmodel.policy import WorldModelPolicy, PlanConfig
from stable_worldmodel.solver import CEMSolver
world = swm.World("swm/PushT-v1", num_envs=8)
world.collect("data/pusht_demo.lance", episodes=100)
solver = CEMSolver(model=your_world_model, num_samples=300)
policy = WorldModelPolicy(solver=solver, config=PlanConfig(horizon=10, receding_horizon=5))
world.set_policy(policy)
results = world.evaluate(episodes=50)
```

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

## Sources

- [stable-worldmodel GitHub repository](https://github.com/galilai-group/stable-worldmodel)
- [Stable World-Model documentation](https://galilai-group.github.io/stable-worldmodel/)
- [stable-worldmodel paper on arXiv](https://arxiv.org/abs/2605.21800v1)