Request to increase the maximum number of test data entries that can be uploaded to the Evaluation Studio feature

[Request]
Because we would like to evaluate with around 4,000 data entries in this project, we kindly request that the upper limit be modified to allow at least 10,000 entries to be processed.

I would greatly appreciate it if you could let me know whether it is possible to proceed, and if so, please share an estimated timeline for when it could be addressed.

[Background]
Our client has been leveraging watsonx.ai for the classification of error messages in system monitoring operations, and since March 2025, it has been in use for production operations.

Currently, we have added watsonx.governance and conducted a PoC with the following intended benefits for operating watsonx.ai:
* Visualization of AI model operations
* Improved efficiency in AI model evaluation and selection
* Minimization of operational costs (Note: This eliminates the need to maintain separate evaluation scripts, as previously required. Additionally, when model parameters or prompts are changed, it removes the need to modify the ServiceNow scripts that call the API.)

The verification results were generally well received, and we are now in discussions with our client regarding a production usage contract.

However, during the verification, we identified an issue that significantly affects the operation of watsonx.governance.
This AHA is a request for improvement regarding that issue.

[Issue and request]
Initially, various documents stated as follows, so we expected that 50,000 records could be uploaded into Evaluation Studio to evaluate foundation models.
However, when about 3,000 records were uploaded into Evaluation Studio, an error occurs with a message indicating that the maximum is 1,000 records.

As a result of confirmation with support, it was clarified that the specifications are as follows:

* Project and pre-production space prompt assets

- Essentials plan: 1000 entries, Lite plan: 500 entries

* Assets in the production space: 3000 entries

In addition, some documents have already been corrected as follows:
"Maximum of 50,000 records per Predictive Model Evaluation
Maximum of 3,000 records per Foundational Model Evaluation”

* Product page: “Maximum 50K records per evaluation”
- (JP) https://www.ibm.com/jp-ja/products/watsonx-governance/pricing
- (EN) https://www.ibm.com/products/watsonx-governance/pricing
* Manual: “Maximum of 50,000 records per evaluation”
- (EN) https://www.ibm.com/docs/en/watsonx/saas?topic=cloud-watsonxgovernance-plans#wos-plan-options-xgov-plans-essential
- (JP) https://www.ibm.com/docs/ja/watsonx/saas?topic=cloud-watsonxgovernance-plans#wos-plan-options-xgov-plans-essential
* Cloud Catalog https://cloud.ibm.com/catalog/services/watsonxgovernance
- (Japanese display) “Maximum 50K records per evaluation”
- (English display) “Maximum 3k records per Foundational Model Evaluation” (-> At that time, only this part was accurate.)

[Impact]
Although it is possible to avoid the problem by splitting the test data, additional tasks will include creating and maintaining a script to split the test data.
Furthermore, since it is necessary to execute the evaluation multiple times, the efficiency of operations will be reduced.