I've migrated all text-related tasks in my automated routines to their API - generation, text checking, various comparisons + the nanobot on the server (mini-openclaw) also works through them.
As you can see from the graph, I've been spending $10 for a week now and it hasn't run out yet :) Similar routines on openai were literally consuming $5-8 per day..
So for now, this is the most cost-effective API without any noticeable loss in quality.
