Vibecoding an inference chip
57 points - today at 6:15 PM
SourceComments
cpldcpu today at 7:51 PM
I did something like this with Opus 4.5: https://github.com/cpldcpu/smollm.c
It seems implementing transformers is the new "hello world" of AI assisted coding.
datadrivenangel today at 7:07 PM
Does not say what type of model or size they were serving, but one of the traces mentions Maple, which is a 20b model. Still, getting ~70+ tok/s is quite good.
codekansas today at 7:30 PM
This is cool
Next step is tuning your post training pipeline to optimize your model for your chosen FPGA architecture...
chalkycrimp today at 7:42 PM
Good chance I'm not the only one thinking this: share the code
spwa4 today at 8:31 PM
I must say I do see one application for this, that you can't really match reasonably otherwise: you can bake weights into an FPGA and then run them for 5W at a speed that will be competitive with a big Nvidia card.
(unfortunately for a decent model it will also be price competitive with a big nvidia card)
saejox today at 7:51 PM
i wonder if we can emulate an FPGA on a rtx 3090? then test it on an AWS EC2 F2 few hours a day.
Archit3ch today at 8:24 PM
Now do silicon. ;)