High-functioning millennial ■ AI 4 XR @ Meta ■ Prev: NVIDIA Research, XR, Omniverse ■ VFX⎯Mathematics⎯Electronic Elsewheres ■ Opinions mine, you can't have them
Tired of your useless GPU hoarding its' precious RAM? Show that bitch who's boss by using VRAM as swap space. Breathe a sigh of relief as the gmail tab takes 15ms less to load
https://t.co/rqypdrHa9e
This is the take I agree with. Memory bandwidth half of M5 Max could matter less depending on how the 128GB unified memory is scheduled.
M1 and GH were both compelling evidence that unified memory solves more pain to devs than it introduces to the market.
I did not expect Nvidia to challenge Apple's unified memory monopoly with a Windows on Arm chip, but the RTX Spark is exactly that. 128GB of LPDDR5X with native CUDA means we can finally run vLLM, Triton, and TensorRT-LLM on a laptop without fighting Metal. This is so good??
In A MATHEMATICIAN'S APOLOGY (1940) G. H. Hardy, depressed from waning physical condition and with a general sense of doom about World War II, writes about the futility of assigning value to a Mathematician's human endeavour.
For over a decade, we’ve accepted that end-to-end backprop is the only way to train deep networks. But holding the entire network in memory all at once is why AI training is hitting a resource wall.
We found a new way to break the network into blocks and train them independently. The trick? Treating the network’s forward pass like a diffusion model denoising a signal.
This reinterpretation slashes the memory needed to train deep models. In our #ICLR2026 paper (https://t.co/PK5h0mqQSo), we matched end-to-end performance across ViTs, DiTs, and LLMs. We did this while training just one isolated block at a time.
@yiningkarlli I initially became a video editor because I wasn't gonna make it as a bass player.
I then did a BSc in Mathematics because the only way to get ahead in TV where I'm from was to wait for someone to retire
I love McMaster-Carr's site as much as the next girl, but if there's *one* site that could use a VLM-backed chatbot, it's def the one you go in to find out if there's a ding-dong strong enough to connect to that whachamacallit under whatever-the-fuck kilonewtons lateral force
I ate one almond the other day. Then I felt bad about the water usage, so I thought to myself - well I just won’t make the next 15,000 AI queries I was planning on making
Most of the furniture I have ever built started with me thrashing around in a ralph loop on https://t.co/FxfUSikwAx until I had a functional CAD of what I'd thought of in the shower just then. There must be a better way and no, it does not involve me reading
@MainzOnX Don't ever yield to the demands of the public. You're here by your actual name and this is real life, no need to change for your audience
(HPC arch)