Jev does seem to have become the Kleenex of decision models. Is brand recognition worth $7.5B? There are lots of other decision models out there that perform at or near jev-level (laya, gliner 2.5 decide, even embedding gemma 2) that you can also run locally, and honestly I think this kind of model makes the most sense running locally as well. Maybe if TypeSafe can ship fast they can stay the default. Guess we'll find out.
I just started experimenting with the decision models. I spun up Laya on a VM with a couple of vCPU and 6GB of RAM. I get the results in about half a second. No need for GPUs or tons of memory.
I am integrating it into the product I am building and to me it doesn't seem like there is much need to go with a SaaS for this since the requirements are so light. I just can run it in Cloud Run and get all of the scale I'll ever need, and I get to tell my customers their data never leaves my environment.