lancedb/docs/web/integrations/stable-worldmodel.mdx

---
title: "Stable World Model"
sidebarTitle: "Stable World Model"
description: "Use Stable World Model with LanceDB-backed datasets for reproducible world model research, fast data loading, and compact storage."
---

import {
  PyFrameworksStableWorldmodelCollectLance,
  PyFrameworksStableWorldmodelConvert,
  PyFrameworksStableWorldmodelEvaluate,
  PyFrameworksStableWorldmodelLoadLance,
} from '/snippets/integrations.mdx';

[Stable World Model](https://github.com/galilai-group/stable-worldmodel) is a research platform for collecting data, training world models, and evaluating policies with model-predictive control across standardized environments.

The LanceDB integration is built into Stable World Model's data format registry. Lance is the default backend for collected datasets, so a path ending in `.lance` gives you an append-friendly LanceDB table with episode-contiguous rows and fast indexed reads.

Random access speed is the bottleneck for world model training, since the loop repeatedly samples temporal windows from high-dimensional observations, actions, and rewards. The faster those windows arrive, the more GPU time goes into training rather than waiting on the data loader.

## Install {#install}

<CodeGroup>
```bash Bash icon="terminal"
pip install stable-worldmodel
```
</CodeGroup>

Datasets and checkpoints are stored under `$STABLEWM_HOME`, which defaults to `~/.stable_worldmodel/`.

## Collect data into Lance {#collect-data-into-lance}

Stable World Model uses Lance by default when you collect to a `.lance` path.
Replace `your_expert_policy` with the expert or scripted policy you use to collect demonstrations.

<CodeBlock filename="Python" language="Python" icon="python">
  {PyFrameworksStableWorldmodelCollectLance}
</CodeBlock>

Every writer accepts a `mode` argument such as `append`, `overwrite`, or `error`. The default is append, so re-running collection extends the existing dataset.

## Load a Lance dataset for training {#load-a-lance-dataset-for-training}

The dataset loader autodetects the Lance format from the path.

<CodeBlock filename="Python" language="Python" icon="python">
  {PyFrameworksStableWorldmodelLoadLance}
</CodeBlock>

Your model code stays focused on the world model objective while LanceDB handles the storage layout and read path.

## Evaluate with model-predictive control {#evaluate-with-model-predictive-control}

After training a world model on the Lance-backed dataset, Stable World Model can evaluate it with planning solvers such as CEM.
Replace `world_model` with the trained model object from your training loop.

<CodeBlock filename="Python" language="Python" icon="python">
  {PyFrameworksStableWorldmodelEvaluate}
</CodeBlock>

## Convert between formats {#convert-between-formats}

Stable World Model can convert between registered dataset formats. A common workflow is to collect in Lance for fast training reads, then export to the video layout for compact inspection artifacts.

<CodeBlock filename="Python" language="Python" icon="python">
  {PyFrameworksStableWorldmodelConvert}
</CodeBlock>

## Throughput {#throughput}

The Stable World Model README reports the following PushT benchmark results from `scripts/benchmark/compare_h5_lance.py`:

| Format | Source | Cache | samples/s | ms/step |
|:---|:---|:---|---:|---:|
| HDF5 | local | no-cache | 1,416.1 | 45.2 |
| HDF5 | local | cached | 1,474.0 | 43.4 |
| LanceDB | local | no-cache | 4,814.8 | 13.3 |
| LanceDB | local | cached | 4,431.3 | 14.4 |
| Video | local | - | 1,330.6 | 48.1 |
| LanceDB | s3 | no-cache | 3,183.7 | 20.1 |
| LanceDB | s3 | cached | 3,253.2 | 19.7 |
| HDF5 | s3 | no-cache | 9.1 | 7,032.5 |
| HDF5 | s3 | cached | 756.5 | 84.6 |

In that benchmark, local LanceDB reached about **3.4x** the no-cache throughput of local HDF5, while S3-backed LanceDB reached about **350x** the no-cache throughput of S3-backed HDF5. Even with cache enabled, S3-backed LanceDB was about **4.3x** faster than S3-backed HDF5.

These numbers come from the Stable World Model project's own benchmark setup, so they're best read as a reproducible directional baseline that may shift across environments, models, and storage configurations.

## Storage {#storage}

The same README reports these local storage sizes for the benchmark dataset:

| Format | Local size |
|:---|---:|
| HDF5 | 43.12 GB |
| LanceDB | 13.31 GB |
| Video | 496.29 MB |

LanceDB used about **69% less local storage than HDF5** in the reported benchmark, while preserving a table interface built for fast training reads and append-heavy collection.

## More resources {#more-resources}

<CardGroup cols={2}>
  <Card title="Stable World Model README" href="https://github.com/galilai-group/stable-worldmodel/blob/main/README.md">
    Installation, quick start, supported formats, benchmarks, environments, solvers, and citation.
  </Card>
  <Card title="Stable World Model docs" href="https://galilai-group.github.io/stable-worldmodel/">
    Full upstream documentation with tutorials, API references, and guides.
  </Card>
</CardGroup>