I don't think your analogy is a good one. Yes, in some way an LLM is a like a Db, but with a fundamental difference: the data is part of the database engine. So when you are the middleman like Stripe/OpenRouter in this case, you are in the middle of a stream of data, which has value on its own - since people pay for it - and some specific characteristics (cost, speed, "intelligence", latency etc) based on which a router can take decisions.
Try this aspect of the analogy:
Anthropic/OpenAI -> Oracle
Gemini -> SQL Server
DeepSeek -> MongoDB
Qwen -> PostgreSQL
Llama -> MySQL
(These are approximate, just as food for thought about broad segments of the markets.)
The important axis in this analogy is that there are likely to be a small number of pure plays that operate as public companies providing LLM services (Anthropic, OpenAI). There will be some companies that sell access to proprietary LLMs as adjacencies to their core product offerings (Gemini). Many/most people will use open offerings without license fees (Llama, Qwen, etc.), paying only for inference (which may be local).
The caveat here is that more pure LLM companies are likely to get public, but I doubt that will be a lasting phenomenon. (Sybase, Informix, Ingres, etc. were also standalone companies at one point, but the market would not support that many commercial database providers.)
Appreciate your taking the time to respond.