Problem

AI Data Platform currently exposes multiple foundation models (e.g. OpenAI, Cohere, Meta Llama), but users are largely left to determine which model is most appropriate for a given task.

This can lead users to follow a trial-and-error approach, resulting in inconsistent results and unnecessary cost. Model selection itself becomes a barrier to adoption rather than an enabler.

Recommendation

Introduce use case driven model guidance, to help users select the right model for the task with confidence.

How it could work is that when a user selects a typical use case (e.g., summarization) the platform could provide clear, opinionated guidance on what available models is best suited for this task, including any key trade-offs (quality, latency, cost)

Example

Use Case: Summarization Task

This approach lets users quickly compare models, understand trade-offs, and make informed choices before running a task, improving adoption and user confidence.