Well, of course you are right in your assertion(s), in a sense. I'm not an expert in Licensing, just a concerned user of tech, but yes, for this exercise I mean GPL, as laid out on the FSF site:
> can you build/setup an AI stack [or environment, or pipeline] that Richard Stallman would use?
GPT4All and RedPajama would both be sufficiently open for him to use, though neither is covered under an FSF-approved license. llama-cpp itself should be perfectly acceptable to him, and there are lots of open API wrappers for these models.
