IBM watsonx

Generate text embeddings using IBM’s watsonx.ai platform.

Supported Models

You can find a list of supported models at IBM watsonx.ai Documentation. The currently supported model names are:

Model ID	Dimensions
`ibm/granite-embedding-278m-multilingual`	768
`ibm/slate-125m-english-rtrvr-v2`	768
`ibm/slate-30m-english-rtrvr-v2`	384
`intfloat/multilingual-e5-large`	1024

For new tables, pass name="ibm/granite-embedding-278m-multilingual". The default name is still the legacy ibm/slate-125m-english-rtrvr, so tables created with it reload with the same model. Legacy and deprecated IDs (such as ibm/slate-125m-english-rtrvr, sentence-transformers/all-minilm-l6-v2 and sentence-transformers/all-minilm-l12-v2) still work for tables whose stored metadata references them, but they aren’t listed above for new use.

Parameters

The following parameters can be passed to the create method:

Parameter	Type	Default Value	Description
name	str	`"ibm/slate-125m-english-rtrvr"`	The model ID of the watsonx.ai model to use. Pass one of the current supported IDs above (e.g. `"ibm/granite-embedding-278m-multilingual"`) when creating new tables.
api_key	str	None	Optional IBM Cloud API key (or set `WATSONX_API_KEY`)
project_id	str	None	Optional watsonx project ID (or set `WATSONX_PROJECT_ID`). Mutually exclusive with `space_id`.
space_id	str	None	Optional watsonx deployment space ID (or set `WATSONX_SPACE_ID`). Mutually exclusive with `project_id`.
url	str	None	Optional custom URL for the watsonx.ai instance
params	dict	None	Optional additional parameters for the embedding model (e.g. `{"truncate_input_tokens": 512}`)

You must supply exactly one of project_id or space_id (either as an argument or via its environment variable). Setting both, or neither, raises a ValueError.

Usage Example

First, the watsonx.ai library is an optional dependency, so must be installed separately:

pip install ibm-watsonx-ai

Optionally set environment variables (if not passing credentials to create directly):

export WATSONX_API_KEY="YOUR_WATSONX_API_KEY"
# Provide exactly one of the following:
export WATSONX_PROJECT_ID="YOUR_WATSONX_PROJECT_ID"
# export WATSONX_SPACE_ID="YOUR_WATSONX_SPACE_ID"

Replace the credential placeholders with your provider credentials. The registry variable keeps the API key out of stored embedding configuration. In application code, load its value from environment variables or a secrets manager.

import tempfile
from pathlib import Path

import lancedb
from lancedb.embeddings import EmbeddingFunctionRegistry
from lancedb.pydantic import LanceModel, Vector

registry = EmbeddingFunctionRegistry.get_instance()
registry.set_var("watsonx-api-key", "your-api-key")
watsonx_embed = registry.get("watsonx").create(
    name="ibm/slate-125m-english-rtrvr",
    api_key="$var:watsonx-api-key",
    project_id="your-project-id",
)

class TextModel(LanceModel):
    text: str = watsonx_embed.SourceField()
    vector: Vector(watsonx_embed.ndims()) = watsonx_embed.VectorField()

data = [
    {"text": "hello world"},
    {"text": "goodbye world"},
]

db = lancedb.connect(str(Path(tempfile.mkdtemp()) / "watsonx-demo"))
tbl = db.create_table("watsonx_test", schema=TextModel, mode="overwrite")
tbl.add(data)

rs = tbl.search("hello").limit(1).to_pandas()
print(rs.head())

Embedding Registry

Embedding Providers

Rerankers

Data Platforms & Frameworks

AI Platforms & Frameworks

Supported Models

Parameters

Usage Example

​Supported Models

​Parameters

​Usage Example

Supported Models

Parameters

Usage Example