MicroLLM lab - tiny LLMs, Q4, in your browser

1 points by dimonomid

Are small language models useful for anything at this time? All the responses I got for simple queries were practically garbage.

dimonomid

I don't think it was meant to be actually practical, no. Just some stuff for fun.
k749gtnc9l3w

At this size, Mozillla's on-device rough translation models are not useless (especially given how cheap running them is). As far as I understand, however, they are seq2seq not completion models.