Benjamin Chetioui

Currently JAX & Pallas @GoogleDeepMind 🇨🇭 (prev XLA @Google 🇨🇭🇧🇻, PL PhD @UiB 🇧🇻) | ⚪⚫ Go player, 🇫🇷🥈'25 | Prev CTF @FlatNetworkOrg. Opinions my own

[Benjamin Chetioui @_SIben_]

@norpadon If Pallas/Mosaic GPU doesn't have what you need, plgpu.inline_mgpu is your escape hatch.

Here is an example of how to use it to insert inline PTX https://t.co/AztFiQOQSV, or arbitrary MLIR code https://t.co/klNSTpggZu.

Please let us know if useful abstractions are missing!

[Adam Paszke @apaszke]

Want to improve GPU compute/comms overlap? We just published a new short tutorial for you!

A few small changes to the Pallas:MGPU matmul kernel is all it takes to turn it into an all-gather collective matmul that overlaps NVLINK comms with local compute: https://t.co/HY4C3MwMb7

[Benjamin Chetioui @_SIben_]

Showcasing one really cool outcome of a small MGPU hackathon we had a few weeks back!

[Adam Paszke @apaszke]

Curious how to write SOTA performance Blackwell matmul kernels using MGPU? We just published a short step-by-step tutorial: https://t.co/XRVX34juEz

At each step, we show exactly what (small) changes are necessary to refine the kernel and the final kernel is just under 150 lines.

[Adam Paszke @apaszke]

@jeremyphoward We have some Pallas tutorials (https://t.co/7LRi5YQLuZ) and in particular a reference guide for Mosaic GPU (https://t.co/jw9lOrjAhA). It is of course far from being comprehensive, but we're working on it. If you have questions you should also just reach out and we'll help :)

[Jacob Austin @jacobaustin132]

Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n

[Benjamin Chetioui @_SIben_]

... And now the thesis is online (https://t.co/lUlxbdCzuE)!

And the defense is scheduled next Wednesday (https://t.co/Q1tFqPx6S0) 😱

It'll in principle be streamed—feel free to DM if you're interested in attending :)

[Benjamin Chetioui @_SIben_]

... And now the thesis is online (https://t.co/lUlxbdCzuE)!

And the defense is scheduled next Wednesday (https://t.co/Q1tFqPx6S0) 😱

It'll in principle be streamed—feel free to DM if you're interested in attending :)

[Benjamin Chetioui @_SIben_]

And almost yet another year, but as of today, the dissertation is submitted! 🥳

[Benjamin Chetioui @_SIben_]

@_xjdr @IlyasHairline @finbarrtimbers FWIW, we have the ability to use Triton to lower a lot of our code in XLA:GPU (and in fact, do it)!

I'm curious exactly what are the performance cliffs that you're facing, and whether we can help with that!

[Benjamin Chetioui @_SIben_]

@Sh0ckFR Merci ! Noté 😉 ce sera avec plaisir

[Benjamin Chetioui @_SIben_]

And almost yet another year, but as of today, the dissertation is submitted! 🥳

[Benjamin Chetioui @_SIben_]

First dissertation draft ready! 🥳 Only took me 5 years and 363 days out of a 4-year contract🤡

[Benjamin Chetioui @_SIben_]

Genuinely excited about this one, can't wait to find out where Mosaic GPU can take H100 performance!

[Adam Paszke @apaszke]

Many of you are excited about H100 attention, so it’s a good time to show you Mosaic GPU: a Python DSL for H100s.

The attention example matches FA3 performance, while being only ~200 lines of Python: https://t.co/12ecz3LftV

It's easy to install too! Latest JAX packages have it.

[Sharad Vikram @sharadvikram]

Finally got around to writing a guide for matrix multiplication on TPUs using Pallas. Check it out!
https://t.co/4RctvwDMWm

[Benjamin Chetioui @_SIben_]

@PatrickKidger Likewise just moved to Zürich! I followed some of your work while at Google, would love to meet up :)

[Daniel Cuthbert @dcuthbert]

As much as I'm personally enjoying the AI ride at the moment, the attacks we are seeing are worryingly familiar web app attacks and one wonders if anyone in the AI world is aware of how you do web app security?

Case in point: https://t.co/fLRVnIUlLT

A very good paper firstly

[shane @shncldwll]

Tired of “LLM hacking” hype with no code? Here’s a breath of fresh air. https://t.co/GAs2gtXz21

1. Challenges: open source ✅
2. Solution framework: open source ✅

If you’re interested in hackbots in offsec and you’re craving something you can RUN, you gotta read this

[Benjamin Chetioui @_SIben_]

@olafurpg Congratulations Olaf! All the best 😊

[Benjamin Chetioui @_SIben_]

@MathisHammel Tu me fais rêver

[Benjamin Chetioui @_SIben_]

@soumithchintala @AlexLaterre @sarahookr Note that JAX/XLA has also gotten *a lot better* on NVIDIA GPUs over the past year (and we're still going) :)

[Flat Network Society @FlatNetworkOrg]

High quality steganographers

[Benjamin Chetioui @_SIben_]

@oas004 @bentrengrove @GoogleOSS Congrats Odin! Great work! :)