Configuring additional models in watsonx BI

You can choose to install other models to change the way that watsonx BI handles certain workloads.

Before you begin

The watsonx BI service must be installed. For more information about the resources that are required for different models, see GPU requirements for models.

About this task

The only models which are required for watsonx BI are granite-3-8b-instruct and ibm-slate-30m-english-rtrvr. You can choose to add the following optional model configurations:
ibm-slate-30m-english-rtrvr on GPU
By default, the ibm-slate-30m-english-rtrvr model runs on CPU. If you have heavy workloads, you can choose to run the ibm-slate-30m-english-rtrvr model on GPU.
OpenAI gpt-oss-120b
The OpenAI gpt-oss-120b model is an optional model. If you want to conduct advanced, configurable reasoning in complex BI queries, you can install the gpt-oss-120b model.
Patch 1 For SQL generation tasks and other language-model-driven functions in watsonx BI, you can use the OpenAI gpt-oss-120b.

Procedure

To install optional models on watsonx BI:

  1. Check which models are installed:
    oc get watsonxaiifm watsonxaiifm-cr \
-n ${PROJECT_CPD_INST_OPERANDS} \
-o jsonpath="{.spec.install_model_list}"
  2. Install the models that you want on watsonx BI:
    • Run the ibm-slate-30m-english-rtrvr model on GPU by modifying the custom resource:
      oc patch watsonxaiifm watsonxaiifm-cr \
--namespace=${PROJECT_CPD_INST_OPERANDS} \
--type=merge \
--patch='{"spec":{"ibm_slate_30m_english_rtrvr_replicas": 1, "ibm_slate_30m_english_rtrvr_resources": {"limits": {"cpu": "2", "memory": "4Gi", "nvidia.com/gpu": 1}, "requests": {"cpu": "1", "memory": "4Gi", "nvidia.com/gpu": 1}}}}'
    • Install the OpenAI gpt-oss-120b model:
      MODEL_LIST="${MODEL_LIST}, \"gpt-oss-120b\""
  3. Update the watsonxaiifm custom resource:
    oc patch watsonxaiifm-cr \
-n ${PROJECT_CPD_INST_OPERANDS} \
-\-type=merge \
-\-patch="{\"spec\":{\"installed_model_list\":[${MODEL_LIST}]}}"
  4. Confirm that the models were added to the spec section of the watsonxaiifm custom resource:
    oc get watsonxaiifm watsonxaiifm-cr \
-n ${PROJECT_CPD_INST_OPERANDS} \
-o jsonpath="{.spec.install_model_list}"

What to do next

If you installed the OpenAI gpt-oss-120b model, select this model by opening watsonx BI, then going to Configuration and settings > Model settings.

After you install and select your models, watsonx BI is ready to use. To begin using watsonx BI, see Getting started with watsonx BI.