noah34

the website is still up, but has been read-only since 2006. the site does not support https, so please use caution.

not surprised about the power user gap section. In my org we have some devs who burn their token allotment in the first day of the month, and never write code by hand, as well as older guys who refuse to touch it.

I always feel like there are gaps between benchmarks and real-world performance on all models, but especially open models as of late. I've used deepseek and kimi (albeit 2.6) extensively and while they work well for maybe 70 percent of tasks, its the last 30 they always trip themselves up on.

They all seem to be very poor at long-running tasks, maintaining context when a change spans multiple areas/layers of a codebase, and making architectural choices. That said, for making the next todo app or ai powered calorie tracker, they are just fine, and the most consumer friendly pricing.

i've been wondering recently if defense against prompt injection is more reliant on system prompt + fine-tuning and reinforcement training, or if it is simply how smart your model is.

to be honest i forgot chromebooks exist and now that you mention them i think they're the best solution for anyone whose computer is simply a medium to access a web browser

i've lost track of which agreement we are on

crazy that they send me emails for all kinds of other bullshit but not this

to many people, Windows, IS "computer"

good luck getting ol' grandma to flash her desktop with Fedora/equivalent

must be all those damn openclaws

good intentions or not, they are letting people have their money back - not that this would put a dent in current enterprise spend anyways

they've been offering refunds for people specifically because of the Fable situation