> ## 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.

# ColPali

We support [ColPali](https://github.com/illuin-tech/colpali) model embeddings for multimodal multi-vector retrieval. ColPali produces multiple embedding vectors per input (multi-vector), enabling more nuanced similarity matching between text queries and image documents.

Using ColPali requires the colpali-engine package, which can be installed using `pip install colpali-engine`.

> [!NOTE]
>
> ColPali produces **multi-vector** embeddings, meaning each input generates multiple embedding vectors rather than a single vector. Use `MultiVector(func.ndims())` instead of `Vector(func.ndims())` when defining your schema.

Supported models are:

* Metric-AI/ColQwen2.5-3b-multilingual-v1.0 (default)
* vidore/colpali-v1.3
* vidore/colqwen2-v1.0
* vidore/colSmol-256M

Supported parameters (to be passed in `create` method) are:

| Parameter | Type | Default Value | Description |
| - | - | - | - |
| `model_name` | `str` | `"Metric-AI/ColQwen2.5-3b-multilingual-v1.0"` | The name of the model to use. |
| `device` | `str` | `"auto"` | The device for inference. Can be `"auto"`, `"cpu"`, `"cuda"`, or `"mps"`. |
| `dtype` | `str` | `"bfloat16"` | Data type for model weights (bfloat16, float16, float32, float64). |
| `pooling_strategy` | `str` | `"hierarchical"` | Token pooling strategy: `"hierarchical"`, `"lambda"`, or `None`. |
| `pool_factor` | `int` | `2` | Factor to reduce sequence length when pooling is enabled. |
| `batch_size` | `int` | `2` | Batch size for processing inputs. |
| `quantization_config` | `Optional[BitsAndBytesConfig]` | `None` | Quantization configuration for the model (requires bitsandbytes). |

This embedding function supports ingesting images as both bytes and URLs. You can query them using text.

import tempfile
from pathlib import Path

import lancedb
import pandas as pd
import requests
from lancedb.embeddings import get\_registry
from lancedb.pydantic import LanceModel, MultiVector

db = lancedb.connect(str(Path(tempfile.mkdtemp()) / "colpali-demo"))
func = get\_registry().get("colpali").create()

class Images(LanceModel):
    label: str
    image\_uri: str = func.SourceField()
    image\_bytes: bytes = func.SourceField()
    vector: MultiVector(func.ndims()) = func.VectorField()
    vec\_from\_bytes: MultiVector(func.ndims()) = func.VectorField()

table = db.create\_table("images", schema=Images)
labels = \["cat", "dog", "horse"]
uris = \[
    "http\://farm1.staticflickr.com/53/167798175\_7c7845bbbd\_z.jpg",
    "http\://farm9.staticflickr.com/8387/8602747737\_2e5c2a45d4\_z.jpg",
    "http\://farm9.staticflickr.com/8216/8434969557\_d37882c42d\_z.jpg",
]
image\_bytes = \[requests.get(uri).content for uri in uris]
table.add(
    pd.DataFrame(\{"label": labels, "image\_uri": uris, "image\_bytes": image\_bytes})
)


Now we can search using text queries:

actual = (
    table.search("a furry pet", vector\_column\_name="vector")
    .limit(1)
    .to\_pydantic(Images)\[0]
)
print(actual.label)


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