Skip to Main Content
IBM Data and AI Ideas Portal for Customers

This portal is to open public enhancement requests against products and services offered by the IBM Data & AI organization. To view all of your ideas submitted to IBM, create and manage groups of Ideas, or create an idea explicitly set to be either visible by all (public) or visible only to you and IBM (private), use the IBM Unified Ideas Portal (

Shape the future of IBM!

We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:

Search existing ideas

Start by searching and reviewing ideas and requests to enhance a product or service. Take a look at ideas others have posted, and add a comment, vote, or subscribe to updates on them if they matter to you. If you can't find what you are looking for,

Post your ideas

Post ideas and requests to enhance a product or service. Take a look at ideas others have posted and upvote them if they matter to you,

  1. Post an idea

  2. Upvote ideas that matter most to you

  3. Get feedback from the IBM team to refine your idea

Specific links you will want to bookmark for future use

Welcome to the IBM Ideas Portal ( - Use this site to find out additional information and details about the IBM Ideas process and statuses.

IBM Unified Ideas Portal ( - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM. - Use this email to suggest enhancements to the Ideas process or request help from IBM for submitting your Ideas.

IBM Employees should enter Ideas at


Showing 22

Start IDE with fraction of CPU request or with no cpu request

Today most IDE in WatsonStudio need to be started with 1 CPU. The customer would like the ability to start IDE with franctions of cpu and not a full cpu or even if possible with no cpu request Customer uses Jupyter, Spark and rStudio IDE. Today us...
17 days ago in 0 Submitted

Cloud pak for Data Watson Studio generating IDE Spark, jupyter, rstudio creates env variables that can affect performance

Customer ran into an issue where all pods in the CPD namespace were slow. A cron job is running in CP4D was suppose to kill idle spark IDE. But there were some pods that were not cleaned. -> With the help o...
17 days ago in 0 Submitted

Spark pods, do better clean up for idle pods

In ticket TS016219727 we did some enhancement to the cronjob that cleans up spark idle jobs. This is an issue with CP4D 4.6 and 4.8 on-prem specifically with the Spark IDE started from Watson Studio. We would like for these enahancements to be in ...
20 days ago in 0 Submitted


Hello, In the IBM Watson AI foundation models' APIs, there is currently no standardized method to submit historical data (several previous interactions with the LLM) so that the model can understand that the conversation pertains to a specific con...
2 months ago in 0 Submitted


Hello, In various providers of generative AI solutions (such as OpenAI, Databricks, etc.), we've observed the option to parameterize API responses to be in streaming mode (True or False). Streaming responses enhance the user experience, as it allo...
2 months ago in 0 Submitted

Support MIG GPUs for foundation model serving

The documentation states that MIG is not supported for foundational model serving in watsonx. Our current infrastructure does not allow us to use GPU passthrough, as our internal infrastructure team relies on vGPU and MIG for availability and easy...
2 months ago in 0 Submitted

Provide for language-specific models across geos (specifically granite-8b-japanese to non-Tokyo regions)

While it is critical to offer language-specific text generation models close to the regions we anticipate will use these the most, for global companies it is important to offer text generation capabilities across languages to multiple geos. For ex...
3 months ago in 0 Submitted

Context Window Length for the CodeLlama-34b-Instruct-hf model

CodeLlama-34b-Instruct-hf Context Window Length is currently 4096, was it originally 16,384? content/wsj/analyze-data/...
3 months ago in 0 Submitted

Context Window Length for the mixtral-8x7b-instruct-v01-q model

I believe the Context Window Length for the mixtral-8x7b-instruct-v01-q model was originally 32,768, but is now limited to 4096. content/wsj/analyze-data/fm-models.html?context=wx&pos=2&locale=en#mi...
3 months ago in 0 Submitted

Known Issues displayed in corresponding windows of the product

Imagine there is a known issue running a model in certain conditions, for version 4.8.3. On the page where you run the model, you have a link to "known issues for 4.8.3". That way, if the model fails, users can see it is a known issue, ...
3 months ago in 0 Submitted