Kai-Hsun Chen
@hsun_kaiSWE @ Anyscale | Apache Submarine PMC member & Committer
Similar User
@_MattJiang_
@Jinghao_J
@xu_dong_sun
@chaitybhandari
@yinfang_chen
@tylergu_jiawei
@CacheMisses
@davidyoung8906
@dplyukhin
@zzjas98
@haoran_qiu98
@Jun_Yang00
@LuoStuart
Happy to see more and more post training libraries built on Ray (OpenRLHF, ChatLearn, veRL)!
Introducing veRL by Bytedance! Flexible & efficient framework for LLM RL/RLHF. Ray-based hybrid programming model. 1.5x - 20x throughput boost. Kudos to @PeterSheng66 and team, & @raydistributed Code: github.com/volcengine/verl Paper: arxiv.org/pdf/2409.19256 #veRL #RL #LLMs
Join @hsun_kai and the @anyscalecompute Team on Tues 10/29 for our next Ray Meetup! 🤝 We'll be talking all things distributed computing and how it's being applied in real-world AI projects. Whether you're new to this or have some experience, there's something for everyone.…
In this episode, guest host and AI correspondent @moficodes interviews @richliaw and @hsun_kai from @anyscalecompute about Ray and KubeRay. Join us to learn how this open source project enables AI workloads on #Kubernetes! kubernetespodcast.com/episode/235-ra…
Welcome @KeertiMelkote!
This is a HUGE update for us! I’ve spent a ton of time with @KeertiMelkote over the past few months and the energy is through the roof. Very very few founders have done what he’s done (starting a business in his garage and scaling it to over $5B in revenue). AI is in its…
1/ 🌟 Continuing with our recent feature updates, we are excited to share that structured logging is now available on Ray & Anyscale, making log management and analysis easier than ever.
I'm thrilled to announce that our OSDI'24 paper "Anvil: Verifying Liveness of Cluster Management Controllers" gets the Jay Lepreau Best Paper Award. Congratulations to my teammates and thanks for the recognition. I will present it in today's "Formal Verification" session. (1/6)
This is for people who want to go deep in the weeds.
Hello everyone 👋 I am seeking support to sustain my open-source work. If you are interested in backing my opensource efforts, please do consider sponsoring me ❤️🙏 github.com/sponsors/ad1992
Pre-training your own diffusion model is a breeze with Ray and Anyscale!🍃 - Leverage diverse GPU types for preprocessing and training. - Fault-Tolerant Training ensures uninterrupted training progress. - Optimized infrastructure and algorithms significantly reduce your costs.
We pretrained a stable diffusion model on 2 billion images for under $40K. Here's what we learned. anyscale.com/blog/scalable-…
Lots of things 😆 There have been a handful of rewrites from scratch over the years with @raydistributed (actually from scratch). Some things that come to mind 💻 Programming languages: First prototype was in Rust. Then C++ (heavy multithreading with gRPC as the core RPC…
Hi Robert, considering all the lessons learned building Ray, are there any changes you would have made back when you first started the project? Personally, I always wondered whether the flexibility of dynamic task graphs would eventually lead to performance bottlenecks.
For orchestrating @raydistributed apps, KubeRay is great to manage everything yourself (& @anyscalecompute if you don't 😉) Now we've added Kueue for adv. scheduling like priority (take resources from low priority tasks) & gang (avoid partial provisions)! cloud.google.com/blog/products/…
Just finished reading Peter Thiel’s “Zero to One” recently. The PayPal Mafia is truly remarkable. Respect 🔥
eBay completed its purchase of PayPal on this day in 2002. PayPal founders & employees went on to launch or fund: Tesla YouTube Facebook SpaceX LinkedIn Uber Square Airbnb Stripe Palantir Affirm Yelp Lyft Pinterest Tumblr Twilio Quora Postmates Eventbrite https://t.co/4Vn9LJcg3n
🎉 Announcing Ray Serve and Anyscale Services general availability! Teams at @LinkedIn, @Samsara, @AntGroup + many more have been using Ray to serve LLMs & multi-modal applications in a flexible, performant and scalable way. Read more about the GA release and how companies have…
New release - SkyPilot v0.4.0! ☸ Kubernetes support - EKS/GKE/On-Prem/... 🔌 Expose ports on clusters 📦 Native container support ☁️ More clouds - Hello, @OracleCloud & @Samsung Cloud! 📘 LLM recipes - vLLM, Llama2, Falcon, LocalGPT Release notes: github.com/skypilot-org/s…
Ray Summit this month will be 🔥🔥 🤯 ChatGPT creator @johnschulman2 🧙♀️ @bhorowitz on the AI landscape 🦹♂️ @hwchase17 on LangChain 🧑🚀 @jerryjliu0 on LlamaIndex 👨🎤 @zhuohan123 and @woosuk_k on vLLM 🧜 @zongheng_yang on SkyPilot 🧑🔧 @MetaAI on Llama-2 🧚♂️ @Adobe on Generative AI in…
This is very impressive. "Introducing Ray Serve dramatically improved our production ML pipeline performance, equating to a ~50% reduction in total ML inferencing cost." 💰💰 @Samsara will be speaking in depth about how they scale AI in a cost-efficient manner at #RaySummit.…
United States Trends
- 1. Bama 14,7 B posts
- 2. Milroe 9.109 posts
- 3. #AEWFullGear 47,2 B posts
- 4. $CUTO 6.648 posts
- 5. Oklahoma 31,3 B posts
- 6. Auburn 11,1 B posts
- 7. Notre Dame 21,5 B posts
- 8. Kansas 37,7 B posts
- 9. Vandy 5.235 posts
- 10. Big 12 1.930 posts
- 11. Sooners 4.050 posts
- 12. Colorado 65,4 B posts
- 13. DeBoer 1.162 posts
- 14. #RollTide 2.439 posts
- 15. Ryan Williams 1.058 posts
- 16. Ospreay 2.471 posts
- 17. #NWSLChampionship 2.252 posts
- 18. #3WiserMen 3.952 posts
- 19. Army 350 B posts
- 20. Heisman 11,9 B posts
Who to follow
-
Yuxuan Jiang
@_MattJiang_ -
Jinghao Jia
@Jinghao_J -
Xudong Sun
@xu_dong_sun -
Chaitanya Bhandari
@chaitybhandari -
Yinfang
@yinfang_chen -
Tyler(Jiawei) Gu
@tylergu_jiawei -
Bogdan "Bo" Stoica
@CacheMisses -
Wei Yang
@davidyoung8906 -
Dan Plyukhin
@dplyukhin -
Zijie Zhao
@zzjas98 -
Haoran Qiu
@haoran_qiu98 -
Jun Yang
@Jun_Yang00 -
Stuart Luo
@LuoStuart
Something went wrong.
Something went wrong.