One of the first things i did was to "gap analysis" with existing libs to see if i could find a niche as I don't want to overlap three.js too much...
The full source on github has a bunch of agent tests that compare running tasks with phoria and three.js. It comes out well, with the agents able to build with both (obviously three because it's TRAINED on that, but I provide enough in llms.txt etc for phoria) and phoria code is ~50% smaller for the same task!
The full source on github has a bunch of agent tests that compare running tasks with phoria and three.js. It comes out well, with the agents able to build with both (obviously three because it's TRAINED on that, but I provide enough in llms.txt etc for phoria) and phoria code is ~50% smaller for the same task!