Hey everyone,
I’ve been rethinking some of my tooling recently, especially around API testing and how it fits into modern ML/AI workflows.
One thing I noticed again is how the Postman free plan has evolved over time. The limitations (especially around collaboration) aren’t new, but they do make it harder to use in small team setups or when you’re working across multiple services.
For simple API testing it still works fine, but once your workflow involves things like:
-
model inference APIs
-
data pipelines
-
multi-service architectures
-
or integrating with ML endpoints
…it starts to feel less flexible unless you move to a paid tier.
So I’ve been exploring alternatives that might fit better with more “AI-native” workflows:
-
OpenAPI + Git-based approaches for versioning endpoints
-
lightweight tools like Bruno or Insomnia for local testing
-
more integrated platforms that combine design, testing, and collaboration
I also tried a couple of newer tools like Apidog, which felt closer to an all-in-one workflow (design + testing + collaboration in one place), but I’m still not sure how it holds up for more complex ML pipelines.
Curious what others here are using in 2026.
Are you still using Postman in your ML / API workflows, or have you moved to something more lightweight or infra-friendly?
Would be great to hear what actually works in practice, especially for model serving and API-heavy setups.