> ## Documentation Index
> Fetch the complete documentation index at: https://docs.lancedb.com/llms.txt
> Use this file to discover all available pages before exploring further.

# GPU Indexing

> Build a vector index faster with GPU acceleration, manually in LanceDB Embedded or automatically in LanceDB Enterprise.

**Applies to:** Embedded ✅, Enterprise ✅

GPU acceleration can reduce the time needed to build vector indexes on large
datasets. Actual build time depends on the index configuration, vector
dimensions, storage throughput and available GPU resources.

> [!NOTE]
>
> **Automatic GPU indexing**
>
> Automatic GPU Indexing is currently only available in [LanceDB Enterprise](https://docs.lancedb.com/enterprise/v1-12-0/index.md#architecture).
> Please [contact us](https://www.lancedb.com/contact) to enable this feature for your deployment.
>
> The vector index is created when you call `create_index`. The backend uses GPU resources
> to build `IVF_PQ` indexes, with the number of GPUs set by the deployment's configuration.
>
> This process is also asynchronous by default, but you can use `wait_for_index` to convert it
> into a synchronous process by waiting until the index is built.

`wait_for_index(...)` waits for the index to exist and for `index_stats(...)` to report `num_unindexed_rows == 0`. It can time out if the table is receiving continuous writes while the build is trying to catch up.

GPU acceleration changes how the vector index is built, not the lifecycle after the build. In the
embedded library, rows appended later still need `optimize()` before they are part of the index; a
deployment adds them in the background -- see [Incremental Indexing](https://docs.lancedb.com/enterprise/v1-12-0/indexing/incremental-indexing.md).
`fast_search()` only searches indexed rows.

## Manual GPU indexing in LanceDB Embedded

You can use the Python SDK to manually create the `IVF_PQ` index on a GPU. You'll need
[PyTorch>2.0](https://pytorch.org/). Note that GPU-based indexing is currently only
supported by the synchronous SDK in LanceDB Embedded.

Specify the values `cuda` or `mps` (on Apple Silicon) for the `accelerator` parameter
to enable GPU training on your device.

### GPU indexing on Linux

```python Python icon="python" theme={"theme":{"light":"vitesse-light","dark":"catppuccin-mocha"}}
table.create_index(
    num_partitions=256,
    num_sub_vectors=96,
    accelerator="cuda",
)
```

### GPU indexing on macOS (Apple Silicon)

```python Python icon="python" theme={"theme":{"light":"vitesse-light","dark":"catppuccin-mocha"}}
table.create_index(
    num_partitions=256,
    num_sub_vectors=96,
    accelerator="mps",
)
```

## Performance considerations

* GPU memory usage scales with `num_partitions` and vector dimensions
* Size GPU memory for the training batches and index configuration
* Monitor memory use while choosing training batch sizes
* Measure build time and recall before changing training settings

## Troubleshooting

If you encounter the error `AssertionError: Torch not compiled with CUDA enabled`,
you need to [install PyTorch with CUDA support](https://pytorch.org/get-started/locally/).

## What to read next

> [card] [Incremental Indexing](https://docs.lancedb.com/enterprise/v1-12-0/indexing/incremental-indexing.md)
>
> Keeping an index up-to-date as new data arrives.

> [card] [Slurm](https://docs.lancedb.com/enterprise/v1-12-0/deployment/slurm.md)
>
> Running these GPU workers alongside training on shared Slurm-scheduled infrastructure.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.