RTK Query: Should I use same query for different endpoints if they all have same data structure?

On our project we use RTK Query and we have the following situation:

On back wed we have endpoints, for simplicity, /api/green/cars/, /api/blue/cars and /api/red/cars that share the same data structure.

My colleagues couldn’t agree whether we should create 3 separate RTK endpoints for them: getGreenCars, getBlueCars, getRedCars with static urls or if we should create one endpoint with a dynamic address:

getCars: build.query({
  query: (color) => ({
    url: `/api/${color}/cars`
    method: 'GET'
  })
})

Those who favor the one endpoint solution say that it will allow us to avoid code duplication.

Those who favor the multiple endpoints solution say that it will keep our code more clear (one rtk endpoint per one back end endpoint) as well as we will be able to implement tags properly (there will be POST, PUT and DELETE too, so we have to keep that in mind).

We are pretty much in a deadlock so I wonder if anyone knows a good piece of theory about this or has experienced something similar so there is an unambiguous decision for this situation.