duckduckgo is an aggregate search engine. They do index certain crowdsourced sites but the rest of their results come from partnerships with other search engines from wiki " and from partnerships with other search engines like Yandex, Yahoo!, Bing and WolframAlpha.[3]"

I have tried to search for where leap2 are getting their results from without finding anything substantial. There is fluff about "living search engine", "social search engine" and a few others. I find no evidence for (or against) that they have their own full Internet search engine.

The problem for new companies trying to take on Google is the immenseness of their infrastructure. Lets say you want to start your own search engine.

You have a great idea, some idea for your crawler, etc. Given that you dont have 10 000s of servers to store and run distributed searches etc you are at a great disadvantage. Can you do it with 5 servers in the cloud? How much space does one complete index of the entire internet take now?

Back when Google started, their hardware was modest.