olivier giroux (@__simt__) profile picture

olivier giroux

@__simt__

Dismantling difficulty in concurrent programming.

 California, USA
Joined July 2014
  • Tweets2.9K
  • Following173
  • Followers3.4K
101Media
Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)Photo by olivier giroux (@__simt__)
Replying to @ernire
I personally think that people overestimate the market potential of hard-to-program computer systems. Most computers have to be programmable to most programmers. Most programs have to be programmed by most programmers. There is a very significant human factor in computing.
4
13
2
100
olivier giroux retweeted
Ronan Collobert (@trebolloc) profile picture
Ronan Collobert
@trebollocOct 15
So proud of the MLX team! Muse Spark Open, on your desk: 816B params at 36+ tokens/sec on 4 Mac Studio computers. We've been hard at work on better M5 GPU Neural Accelerator support, a smoother Thunderbolt experience (including multi-cable meshes), and kernel optimizations for MoE architectures. Oh — and MLX LM (mlx_lm.generate, mlx_lm.server) now supports the vision modality for Muse Spark Open! More on using MLX for distributed inference of frontier-scale models on Mac Studio: machinelearning.apple.com/re…
Image from Ronan Collobert (@trebolloc)
John Ternus (@johnternus) profile pictureJohn Ternus
@johnternusOct 15
Replying to @finkd
Congrats to the team! Amazing to run a model of Muse Spark Open's caliber and scale this efficiently, right from your desk. Mac Studio keeps pushing the boundaries of on-device AI!
22
34
10
273
20,886
Never thought I would say this but I think I agree.
Scott (@scottstts) profile pictureScott
@scottsttsSep 25
My god this is such a good speech that every SWE needs to hear. You know what? Every person should hear it Keep the happy memories, eyes on the reality, be excited about the future. That’s the best that anyone can do
2
16
2,315
olivier giroux retweeted
@cdleary (@cdleary) profile picture
@cdleary
@cdlearyAug 30
it is quite analogous to not knowing exactly what assembly comes out of the compiler at -O3 and why, even if you wrote the original "functional spec" of C++ code that fed the compiler
3
2
13
931
Every time I see one of these posts, I go read what the class teaches kids about thread scheduling. This one is ok.
Vivek Galatage (@vivekgalatage) profile pictureVivek Galatage
@vivekgalatageMar 25
Roadmap: Understanding GPU Architecture from Cornell cvw.cac.cornell.edu/gpu-arch…
Image from Vivek Galatage (@vivekgalatage)
Image from Vivek Galatage (@vivekgalatage)
4
17
310
27,649
Hey @united it would be cool if cabin announcements were live-captioned on the seat display.
6
1,334
olivier giroux retweeted
Gokhan Avkarogullari (@gavkar) profile picture
Gokhan Avkarogullari@gavkarMar 20
We brought significant architectural advancements and feature set improvements to A19/M5 GPUs. Scalable GPU Neural Accelerators 2nd Gen Dynamic Caching Shader Architecture 3rd Gen Ray Tracing The best GPU Driven Pipeline New graphics features, rate and perf increases.
17
30
1
577
37,659
Hey, @trebolloc @awnihannun, I wonder if you know of some member of the MLX community who would like to work on neural accelerators directly.
3
1,221
olivier giroux retweeted
Awni Hannun (@awnihannun) profile picture
Awni Hannun
@awnihannunMar 3
M5 Max is a local AI powerhouse in a laptop form factor. So awesome to see this thing released. Up to 8x faster prefill / image generation compared to M1 Max. Benchmarks done with MLX / mlx-lm.
Image from Awni Hannun (@awnihannun)
31
43
10
481
37,076
Kinda curious what Bryce is going to say.
C++ Under the Sea (@cppunderthesea) profile pictureC++ Under the Sea@cppunderthesea21 Aug 2024
Join @blelbach for his talk on the C++ Execution Model. The heart of C++ is a multi-threaded abstract machine & formal model that describe how programs run. Rules govern how code is executed, which impacts your code. Early-bird tickets: store.ticketing.cm.com/cppun… #cpp #cplusplus
Image from C++ Under the Sea (@cppunderthesea)
2
2
22
2,920
BREAKING: P2300 has been voted into C++26! 🎉
15
26
1
175
19,117
olivier giroux retweeted
Randall Munroe (@xkcd) profile picture
Randall Munroe
@xkcd15 May 2024
Driving PSA xkcd.com/2932
Image from Randall Munroe (@xkcd)
142
1,474
161
16,382
530,049
olivier giroux retweeted
Longhorn (@never_released) profile picture
Longhorn@never_released12 Nov 2023
An actual nice excuse to get a new MacBook is the new GPU architecture. It looks really nice
1
35
7,498
This is a great introduction to the new GPU architecture in M3 and A17 Pro SOCs: developer.apple.com/videos/p…
Explore GPU advancements in M3 and A17 Pro - Tech Talks - Videos - Apple Developer

Explore GPU advancements in M3 and A17 Pro - Tech Talks - Videos - Apple Developer

Learn how Dynamic Caching, the next-generation shader core, hardware-accelerated ray tracing, and hardware-accelerated mesh shading of...

developer.apple.com
1
15
51
7,292
olivier giroux retweeted
Gokhan Avkarogullari (@gavkar) profile picture
Gokhan Avkarogullari@gavkar9 Nov 2023
If you re curious about what the new #a17 and M3 GPU and #dynamiccache is about and the brand new and exceptional GPU dev tooling and best shader practices we released new tech talks today: developer.apple.com/videos/p…, developer.apple.com/videos/p…, and developer.apple.com/videos/p…
Explore GPU advancements in M3 and A17 Pro - Tech Talks - Videos - Apple Developer

Explore GPU advancements in M3 and A17 Pro - Tech Talks - Videos - Apple Developer

Learn how Dynamic Caching, the next-generation shader core, hardware-accelerated ray tracing, and hardware-accelerated mesh shading of...

developer.apple.com
10
64
3
249
46,312
It's been a long & satisfying journey to build this brand new scary fast GPU: never before seen architecture, HW ray tracing & mesh shading. Congratulations to all teams at Apple for making this great GPU, great M3 SoCs and great MacBook Pro products that Apple introduced today.
Image from Gokhan Avkarogullari (@gavkar)
17
27
5
404
81,780
More gigantic congrats to #GPU SW @ #Apple! Want to work with us on the latest #m3 family of GPUs for #applesilicon? If yr a C/C++ expert w demonstrable #graphics experience, DM me. #raytracing #meshshading #rendering #m3pro #applejobs #gamedevjobs #gamejobs #m3max #AppleEvent
1
7
25
4,027
Is it just me or the Ubuntu kernels PPA is down? This: kernel.ubuntu.com/~kernel-pp…
4
4
1,836
I haven’t said anything on this hell site in weeks but I agree so much with this.
@fclc cmp lea char (@FelixCLC_) profile picture@fclc cmp lea char@FelixCLC_3 Oct 2023
I'm starting to believe that if NV released GraceHopper in a cutdown WS version, it would become the default Linux ML workstation within a few quarters. Partially because SW stack/CUDA, also because CPU would be powerful enough to do real WS task/be a responsive dev experience.
1
12
2,355
Continuing attempts to escape from here; trying the squirrelly @ sign app. Hopefully it gets critical mass. Because I was so, so very late to join the other app, the vanity photo one, I'm __simt_arch__ over there. 🤞
1
1
5
1,450
Wtf, geez, $19/month now.
Image from olivier giroux (@__simt__)
1
1
1,639
Load more