Sander Dieleman (@sedielem) 's Twitter Profile
Sander Dieleman

@sedielem

Research Scientist at Google DeepMind (WaveNet, Imagen 3, Veo). I tweet about deep learning (research + software), music, generative models (personal account).

ID: 2902658140

linkhttps://sander.ai calendar_today02-12-2014 18:02:01

2,2K Tweet

59,59K Followers

1,1K Following

justin (@honestblogging) 's Twitter Profile Photo

A college professor doing a class on Gen Z slang and the video pans over to all the boomers taking notes and seeming super interested #veo3

Yulia Rubanova (@yuliarubanova) 's Twitter Profile Photo

Introducing Veo Ingredients-to-Video on labs.google/flow Create customized videos from reference images: of people, characters, objects, background, clothing, textures or anything you want! Examples in the thread 🧵

Josh Woodward (@joshwoodward) 's Twitter Profile Photo

Veo 3 dropped about 100 hours ago, and it's been on 🔥🔥🔥 ever since Now, we’re excited to announce: + 71 new countries have access + Pro subscribers get a trial pack of Veo 3 on the web (mobile soon) + Ultra subscribers get the highest # of Veo 3 gens w/ refreshes How to try

Sander Dieleman (@sedielem) 's Twitter Profile Photo

In the visual domain, diffusion is approximate spectral autoregression. Does it have to be? 🤔 Fabian Falck wrote a thought-provoking response to my blog post on the topic, in the form of another blog post.

Katie Everett (@_katieeverett) 's Twitter Profile Photo

There were so many great replies to this thread, let's do a Part 2! For scaling laws between loss and compute, where loss = a * flops ^ b + c, which factors change primarily the constant (a) and which factors can actually change the exponent (b)? x.com/_katieeverett/…

Sander Dieleman (@sedielem) 's Twitter Profile Photo

I once discovered I'd been training networks without any biases for 3 months, because I forgot y += b in my conv layer implementation 🙃 Turns out it didn't really matter 🤷 although that wasn't quite as well-established at the time, so it was a bit of a shock to find out!

Sander Dieleman (@sedielem) 's Twitter Profile Photo

I just had to prove my UK immigration status at the airport for the first time (heading home from SF), using the digital-only system. The people at the check-in desk and I were taken aback by how unintuitive and cumbersome this process is. Physical proof is sorely needed!

Sander Dieleman (@sedielem) 's Twitter Profile Photo

If you've read my latest blog post on generative modelling in latent space, this one is a great follow-up about putting things into practice. openworldlabs.ai/blog/training-…

DrMachakil (@drmachakil) 's Twitter Profile Photo

Yesterday, I played with Google Veo 3, and honestly… the possibilities blew my mind. Here’s a short mockumentary I made with it called “The Prompt Floor”, a behind the scenes look at how AI videos get made from the inside. #googleveo3 #aivideo

Google UK (@googleuk) 's Twitter Profile Photo

We’ve been using Veo 3 to ask Britain's wildlife about their eating habits 🍔🥬. We're excited to see how you're transforming prompts into productions with Gemini! 🎬

Sander Dieleman (@sedielem) 's Twitter Profile Photo

This work uncovers a profound connection between continuous and discrete (non-absorbing) diffusion models, allowing transfer of advanced techniques such as consistency distillation to the discrete setting! Also: amazing title, no notes! 🧑‍🍳😙🤌

Chris Donahue (@chrisdonahuey) 's Twitter Profile Photo

Excited to announce 🎵Magenta RealTime, the first open weights music generation model capable of real-time audio generation with real-time control. 👋 **Try Magenta RT on Colab TPUs**: colab.research.google.com/github/magenta… 👀 Blog post: g.co/magenta/rt 🧵 below

Black Forest Labs (@bfl_ml) 's Twitter Profile Photo

High quality image editing no longer needs closed models We release FLUX.1 Kontext [dev] - an open weights model for proprietary-level image editing performance. Runs on consumer chips. ✓ Open weights available ✓ Best in-class performance ✓ Self-serve commercial licensing

High quality image editing no longer needs closed models

We release FLUX.1 Kontext [dev] - an open weights model for proprietary-level image editing performance. Runs on consumer chips.

✓ Open weights available
✓ Best in-class performance
✓ Self-serve commercial licensing