Jean-Baptiste Dalido
@jbdalidoEngineering @zml_ai, fast motorcycles enthusiast
Similar User
@PluClaire
@corentinanjuna
@steeve
@speude
@michaeljames
@yim1n
@cdethiersant
@julienmartin_
@dahbiahmed
@itsjackiehu
@vbatoufflet
@mc_rollin
@qperez42
@EtaixJ
@Codhex
#Zigtoberfest 2024 is over, and it was an absolute blast! I got to present @zml_ai, @croloris talked about Zine, David Sugar educated us on secure PW DBs, and we also had CHR programming and embedded mem-mapped IO. Pretzels and drinks for free, nobody had to starve. I met so…
Just realized it was defaulting to CPU! While the Intel chips are pretty fast, we get much better performance on the GPU... lol.
Oops, I did it again.
Airbuds engineer extraordinaire, Gawen, is celebrating our 3M user milestone by dyeing his hair 👀❤️💙💜💚🧡 Can you guess the color???
Join us!
Been waiting for this moment and great to see @zml_ai come out of stealth! Congratulations team! 👏🏼 #opensource #GenAI
ZML: a high-performance AI inference stack that can parallelize and run deep learning systems on lots of different hardware. It's out of stealth, impressive, and open source.
ZML: a high-performance AI inference stack that can parallelize and run deep learning systems on lots of different hardware. It's out of stealth, impressive, and open source.
Excited to see this in the wild. In the coming years none of us will be thinking about what hardware we're running.
We are live.
Very proud to release what we've been working on @zml_ai in the past few months !
Dear folks, what are your average tok/s on Llama-3.1-8B BF16 on A100 and H100 for batch_size = 1 ?
BYOC is inevitable like I said many years ago. At the end of the day, Capitalism is always going to win. If you build a product people want, that's what will dominate, everything else is noise. The only thing that matters is happy customers in production. When we launched…
New blog post! A commentary on BYOC, the Warpstream acquisition and Confluent as an everywhere platform. jack-vanlightly.com/blog/2024/9/11…
Sex is good but have you ever packaged and deployed a CUDA binary in Distroless completely self contained ?
So missed this one <3 github.com/redpanda-data/… @redpandadata
Hey folks, since we've been asked so many times, here is a quick demo of what we're building. This is a small LLaMA2 sharded on 1 NVIDIA RTX 4090 (in Paris), 1 AMD 6800XT (in @TheCorendos's flat) and 1 Google Cloud TPU v2 over @Tailscale Exact same code, all built on my Mac
You may not like it but this is what peak Cybertruck looks like @elonmusk
United States Trends
- 1. Mika 54,2 B posts
- 2. #IDontWantToOverreactBUT N/A
- 3. Morning Joe 37,6 B posts
- 4. DeVito 14,3 B posts
- 5. $CUTO 13,7 B posts
- 6. #TSTTPDHolidayEssentials 1.332 posts
- 7. Drew Lock 2.203 posts
- 8. #MondayMotivation 17,8 B posts
- 9. Tommy Cutlets N/A
- 10. Victory Monday 3.204 posts
- 11. Spirit Airlines 1.237 posts
- 12. Good Monday 54,3 B posts
- 13. Ratings 26,7 B posts
- 14. Standout 6.526 posts
- 15. #MondayVibes 3.280 posts
- 16. Joe Scarborough 27,2 B posts
- 17. DeFi 131 B posts
- 18. TABLO 14 B posts
- 19. JUST ANNOUNCED 35,6 B posts
- 20. #18Nov 3.898 posts
Who to follow
-
Claire Pluvinage
@PluClaire -
Corentin Kerisit
@corentinanjuna -
Steeve Morin
@steeve -
spud
@speude -
Michael Goldenstein
@michaeljames -
Yi
@yim1n -
Charlotte de Thiersant
@cdethiersant -
Julien Martin
@julienmartin_ -
Ahmed Dahbi
@dahbiahmed -
jackie hu
@itsjackiehu -
Vincent Batoufflet
@vbatoufflet -
Marc Rollin
@mc_rollin -
Quentin Perez 🦀🧨
@qperez42 -
Jonathan Etaix
@EtaixJ -
Clément Dumas
@Codhex
Something went wrong.
Something went wrong.