haven't heard of it, enjoying my $10 OpenCode Go sub, dual OEM spark setup, and open weights. I also use Fireworks at work.

Every token vendor seems to be having issues, by proxy of hyperscaler capacity issues. Whether or not we get served up quants is opaque, and should be a finable offense for not delivering what you are charging for.

The dual spark is definitely nice when this happens, and even though I doubt I'll ever pay it back in terms of saved token costs, the freedom sure does feel great!