Ultrafast machine learning on FPGAs via Kolmogorov-Arnold Networks (aarushgupta.io)
92 points by ag2718 2 hours ago
mikeayles 27 minutes ago
So for people wondering if it can be used to accelerate LLM inference, sadly not.
I've been trying to hit 100,000tokens/s with a 3.28m dumb model, and even this is an order of magnitude too large to benefit.
It appears to be focussed more on latency, than throughput. Happy to be corrected?
ag2718 23 minutes ago
You're correct that this work is not very applicable for LLMs and that the focus here is primarily on latency.
tomrod 20 minutes ago
Happy to hear that KANs continue to find solid footing.
RantyDave 2 hours ago
Right. But ... this would limit you to either extremely small models or extremely large FPGA's, yes? If there's a simple machine learning task that requires a sub microsecond latency I can see the point but otherwise??
ag2718 an hour ago
Yes, this work is focused on accelerating very small models, typically for real-time systems that require extremely low power or low latency.
One primary application of this work is in high-energy physics (https://home.cern/smarter-decisions-at-the-speed-of-collisio...). Ultrafast and real-time learning is also very applicable for problems in quantum computing, plasma control, etc. (https://arxiv.org/pdf/2602.02005).
poly2it an hour ago
I'm not in HFT, but I assume this is also an interesting applicable domain?
ag2718 an hour ago
Animats an hour ago
This guy will be hired by a high-frequency trading firm, and the next time we hear about him, he will have a net worth in 9 figures.
throwaw12 an hour ago
he is already at Jane Street
Animats an hour ago
Of course.
babelfish an hour ago
Archive link, as it looks like the original post was taken down: https://web.archive.org/web/20260609200156/https://aarushgup...
ag2718 an hour ago
Hmm the post is still up for me?
dang 43 minutes ago
For us too, but we'll put the archive link in the toptext since these things seem to vary a lot by region.
p.s. Thanks for posting this and welcome to HN!