langchain-ibm

This package provides the integration between LangChain and IBM watsonx.ai through the ibm-watsonx-ai SDK.

Installation

To use the langchain-ibm package, follow these installation steps:

pip install -U langchain-ibm

Setting up

To use IBM's models, you must have an IBM Cloud user API key. Here's how to obtain and set up your API key:

Obtain an API Key: For more details on how to create and manage an API key, refer to IBM's documentation.
Set the API Key as an Environment Variable: For security reasons, it's recommended to not hard-code your API key directly in your scripts. Instead, set it up as an environment variable. You can use the following code to prompt for the API key and set it as an environment variable:

import os
from getpass import getpass

watsonx_api_key = getpass()
os.environ["WATSONX_API_KEY"] = watsonx_api_key

In alternative, you can set the environment variable in your terminal.

Linux/macOS: Open your terminal and execute the following command:
```
export WATSONX_API_KEY='your_ibm_api_key'
```
To make this environment variable persistent across terminal sessions, add the above line to your ~/.bashrc, ~/.bash_profile, or ~/.zshrc file.
Windows: For Command Prompt, use:
```
set WATSONX_API_KEY=your_ibm_api_key
```

Setting parameters

You might need to adjust model parameters for different models or tasks. For more details on the parameters, refer to Parameter Scheme IBM's documentation.

Note: You must use the correct parameter schema for the class you are initializing:

ChatWatsonx (for chat) uses TextChatParameters
WatsonxLLM (for text generation) uses TextGenParameters.
WatsonxRerank (for reranking) uses RerankParameters.

This example uses ChatWatsonx, so we import TextChatParameters.

from ibm_watsonx_ai.foundation_models.schema import TextChatParameters

parameters = TextChatParameters(
    temperature=0.5,
    max_completion_tokens=1024,
    top_p=1,
)

You can also pass it as a dictionary object.

parameters = {
    "temperature": 0.5,
    "max_completion_tokens": 1024,
    "top_p": 1,
}

Chat Models

ChatWatsonx class exposes chat models from IBM.

Initialization the ChatWatsonx class with the previously set parameters.

from langchain_ibm import ChatWatsonx

model = ChatWatsonx(
    model_id="PASTE THE CHOSEN MODEL_ID HERE",
    url="PASTE YOUR URL HERE",
    project_id="PASTE YOUR PROJECT_ID HERE",
    params=parameters,
)

model.invoke("Sing a ballad of LangChain.")

Note:

You must provide a project_id or space_id. For more information refer to IBM's documentation.
Depending on the region of your provisioned service instance, use one of the urls described here.
You need to specify the model you want to use for inferencing through model_id. You can find the list of available models here.

Alternatively for all classes you can use Cloud Pak for Data credentials. For more details, refer to IBM's documentation.

from langchain_ibm import ChatWatsonx

model = ChatWatsonx(
    model_id="ibm/granite-4-h-small",
    url="PASTE YOUR URL HERE",
    username="PASTE YOUR USERNAME HERE",
    password="PASTE YOUR PASSWORD HERE",
    project_id="PASTE YOUR PROJECT_ID HERE",
    params=parameters,
)

Embedding Models

WatsonxEmbeddings class exposes embeddings from IBM.

from langchain_ibm import WatsonxEmbeddings
from ibm_watsonx_ai.metanames import EmbedTextParamsMetaNames

embed_params = {
    EmbedTextParamsMetaNames.TRUNCATE_INPUT_TOKENS: 3,
    EmbedTextParamsMetaNames.RETURN_OPTIONS: {"input_text": True},
}

embeddings = WatsonxEmbeddings(
    model_id="ibm/granite-embedding-107m-multilingual",
    url="https://us-south.ml.cloud.ibm.com",
    project_id="PASTE YOUR PROJECT_ID HERE",
    params=embed_params,
)

embeddings.embed_query("What is the meaning of life?")

LLMs

WatsonxLLM class exposes LLMs from IBM.

from langchain_ibm import WatsonxLLM
from ibm_watsonx_ai.foundation_models.schema import TextGenParameters, TextGenDecodingMethod

parameters = TextGenParameters(
    decoding_method=TextGenDecodingMethod.SAMPLE,
    temperature=0.5,
    top_k=50,
    top_p=1
)

llm = WatsonxLLM(
    model_id="ibm/granite-4-h-small",
    url="https://us-south.ml.cloud.ibm.com",
    project_id="PASTE YOUR PROJECT_ID HERE",
    params=parameters,
)

llm.invoke("The meaning of life is")

Reranker

WatsonxRerank class exposes reranker from IBM.

from langchain_ibm import WatsonxRerank

rerank = WatsonxRerank(
    model_id="cross-encoder/ms-marco-minilm-l-12-v2",
    url="https://us-south.ml.cloud.ibm.com",
    project_id="PASTE YOUR PROJECT_ID HERE",
)

Toolkit

WatsonxToolkit class exposes Toolkit from IBM.

from langchain_ibm.agent_toolkits.utility import WatsonxToolkit

watsonx_toolkit = WatsonxToolkit(
    url="https://us-south.ml.cloud.ibm.com",
)

Metadata

Release files for langchain-ibm 1.1.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for langchain-ibm 1.1.2
File	Size	Uploaded
langchain_ibm-1.1.2.tar.gz	75.1 kB	Oct 5, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for langchain-ibm 1.1.2
File	Interpreter	ABI	Platform	Reset
langchain_ibm-1.1.2-py3-none-any.whl 54.8 kB Oct 5, 2026	Python 3	none	any	Details

Total release size: 129.9 kB

Release files / langchain_ibm-1.1.2.tar.gz

Download URL	langchain_ibm-1.1.2.tar.gz
Size	75.1 kB
Tags	Source
SHA-256 checksum How to use checksums	`ae162e7a8eaaf72a945abc92609638049619691e3f4a7c338b80a360574f0bbc`
BLAKE2b-256 checksum How to use checksums	`4277f1729121f208ddfeae5995631ebd4011a8a7f9dee81456396548a303d23d`
Upload date	Oct 5, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Release files / langchain_ibm-1.1.2-py3-none-any.whl

Download URL	langchain_ibm-1.1.2-py3-none-any.whl
Size	54.8 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`bf62265f8734072746e6e804e9af73d934fb578fedc0388b81778d45352642b0`
BLAKE2b-256 checksum How to use checksums	`cf0a1afc846f961f1509f30b685babf19e62e6ebfb9822fdf200696aeb33fd0a`
Upload date	Oct 5, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`