Doing it from the mobile will always be slower than Postman. because the mobile will call the Mendix server and the Mendix server will call the API then will return the results.
Postman → API Server
Postman ← API Server
Mobile → Mendix Server
Mendix Server → API Server
Mendix Server ← API Server
Mobile ← Mendix Server
Also have a look at the logic that you implemented in the Nanoflow usually mobile phones are slow in processing the data, if this data doesn’t change frequently then try to call the API fewer times and when the user needs the data just get it from the Mendix server.
The increased time in mendix could be attributed to the size and complexity of the returned objects. The rest call in mendix not only retrieves data but also parses it and create internal mendix objects for further processing.
whether that is the root cause here or not is hard to conclude without example of the result from postman.