bytesandbits

8 ms·

1.
by bytesandbits 5d ago
This is just bait for regulatory capture. It’s the only way in which you can find alignment of Elon, Sam and Dario within the same 24h.
2.
by bytesandbits 5d ago
Zoom is malware, has always been.
3.
by bytesandbits 1mo ago
No Small Actors.
4.
by bytesandbits 2mo ago
I imagine it is based off World Labs
5.
by bytesandbits 2mo ago
absolutely crazy level of detail... in awe
6.
by bytesandbits 5mo ago
I thank Microsoft deeply for the forced copilot crap, almost impossible to remove, that they have put into vs code. Finally after 5 years I have deleted vs code in my Mac! that was the last piece of windows software I still had around. VS c
7.
by bytesandbits 5mo ago
I know a 10x engineer when i see one.
8.
by bytesandbits 5mo ago
fine. I will port it myself. Real-time, sub 100ms latency. Here https://testflight.apple.com/join/myNP5XvU
9.
by bytesandbits 5mo ago
Handy with parakeet is pretty awesome by the way!
10.
by bytesandbits 6mo ago
here the original NASA photos at high resolution without unnecessary ads. https://www.nasa.gov/gallery/journey-to-the-moon/
11.
by bytesandbits 6mo ago
why is a mail client needed in an onboard space computer at all ?
12.
by bytesandbits 6mo ago
anything that doesn't touch the model parameters at all once it has been compiled. for example, in streaming ASR of an encoder-decoder you can get gains in accuracy just by enhancing the encoder-decoder orchestration and ratio, frequen
13.
by bytesandbits 6mo ago
we constantly underestimate the power of inference scaffolding. I have seen it in all domains: coding, ASR, ARC-AGI benchmarks you name it. Scaffolding can do a lot! And post-training too. I am confident our currently pre-trained models c
14.
by bytesandbits 6mo ago
https://machinelearning.apple.com/research/exploring-llms-ml...
15.
by bytesandbits 6mo ago
Apple
16.
by bytesandbits 7mo ago
4x faster PREFILL not decode. Decode is bandwidth-bounded. Prefill is flops-constrained.
17.
by bytesandbits 7mo ago
do you run two eSIMs when traveling and if so how is stability / battery life?
18.
by bytesandbits 7mo ago
incredible work
19.
by bytesandbits 7mo ago
I am not sure what "edge" device you want to run this on, but you can compress parakeet to under 500MB on RAM / disk with dynamic quants on-the-fly dequantization (GGUF or CoreML centroid palettization style). And retain esse
20.
by bytesandbits 7mo ago
sensei karpathy has done it again
21.
by bytesandbits 7mo ago
parakeet v3 has a much better RTFx than moonshine, it's not just about parameter numbers. Runs faster. https://huggingface.co/spaces/hf-audio/open_asr_leaderboard
22.
by bytesandbits 7mo ago
maybe a deepseek v4 distill. give it a few days
23.
by bytesandbits 7mo ago
its cause of a chain of events. Next week Chinese New year -> Chinese labs release all the models at once before it starts -> US labs respond with what they have already prepared also note that even in US labs a large proportion of re
24.
by bytesandbits 7mo ago
Not trained in Ascend that is BS. Hopper GPU cluster. Please remove that.
25.
by bytesandbits 8mo ago
wow Mistral really cooked
26.
by bytesandbits 8mo ago
how are you so fast man
27.
by bytesandbits 8mo ago
John Giannandrea leaves and things start to work well in the Apple AI department. Maybe just a coincidence :)
28.
by bytesandbits 8mo ago
I will never understand why when you expect to sell a few hundred units of a product at most, you would decide to go and build something in china. There are instances where building in china might make sense. this is not one of them! you
29.
by bytesandbits 8mo ago
imbue
30.
by bytesandbits 8mo ago
great work! I am going to try it out. Currently about to learn some Mandarin to be able to talk with hawker stand owners for a trip I am doing soon. I am trilingual and can speak a few languages on top of that, but none of them tonal. I am

More ›