I like my Apache 2.0 licensed Gemma, and NVIDIA’s Nemotrons are decent bases for finetuning or continued pretraining, esp thanks to good documentation and tooling.
replyOh, and Mira’s thinking machines lab dropped Inkling, a ~1T open weight model too.
This isn’t US vs China. This is open vs closed.
This isn’t open though. Promises to be open later aren’t worth anything, given what we’ve seen and heard from AI execs making promises in this industry.
replyit takes extra effort to open source a model even if you had it running internally. even traditional software takes extra effort to get released as open source.
replyI’m sympathetic but AI claims are to be disbelieved until proven anymore. That’s a well poisoned thoroughly by not just American companies.
replyIt's Sunday morning so I'm allowed to be facetious!
reply