NVFP4, every mention
4 scenes across 3 shows · ← back to NVFP4
tap a year for its mentions
the MAD Podcast 5
Latent Space 2
the Y Combinator Startup Podcast 1
every year every show the MAD Podcast 5 Latent Space 2 the Y Combinator Startup Podcast 1
Verbatim, from the transcripts: passages where NVFP4 comes up on the MAD Podcast, Latent Space, the Y Combinator Startup Podcast
Next 100x in AI: Inference, Networking, & Self-Optimizing Models — Philip Kiely & Ali Taha, Baseten
- ▶ 1:01:12 Philip Kiely NVFP four is big. 2 times in the scene
Multi-GPU Kernels, Intelligence per Watt, Heterogeneous Inference, and More | YC Paper Club · Y Combinator
- ▶ 35:11 unnamed speaker And then, uh, starting this January, you know, we did this, like, very, like, you know, prolific, like, this very popular competition on NVFP four kernels, and people who've never written GPU kernels before started to get really…
Inside Nemotron & NVIDIA’s AI Lab | Bryan Catanzaro
- ▶ 35:48 Bryan Catanzaro We pre-trained those in MVFP four, um, which, you know, ah, is, ah, a not trivial thing to do, to invent the algorithm so that your model can converge to an excellent result using such coarse arithmetic, ah, required a lot of invention. 2 times in the scene
- ▶ 1:09:38 Bryan Catanzaro Um, uh, NVFP for pre-training is one of them. 3 times in the scene