A distributed compute framework for scaling AI workloads. Created and developed by @anyscalecompute.
Joined August 2019
- Tweets2K
- Following2
- Followers11.7K
- Likes2K
ray retweeted
Kubernetes doesn’t know what a Ray cluster is. So teams keep hand writing the same Deployments, Services, and recovery scripts to run Ray on it.
KubeRay fixes that: autoscaling, fault tolerance, and zero downtime upgrades for Ray, built into Kubernetes.
na2.hubs.ly/H08qhRS0
ray retweeted
@Xoople processes 500,000 km² of satellite imagery in under 5 minutes on 12 A10 GPUs with Anyscale on @Azure.
A single request can fan out to 2 trillion pixels. 15 engineers without a dedicated infra team.
Read the case study: na2.hubs.ly/H08qgTD0.
ray retweeted
Run Reflection's Beam at scale in your own account for full sovereignty but without managing infra with Ray Serve LLM on Anyscale!
Excited to announce Day 0 support for Reflection's Beam model on Ray Serve LLM!
Beam is Reflection's first open-weight model, advancing the frontier of Western AI in the 500B-parameter class. Built as a sparse MoE model, Beam is optimized for coding, agentic workloads, and efficient inference.
We've validated Ray Serve LLM deployment recipes for Beam along with its DFlash2 drafter. 🧵
ray retweeted
Open Source AI Week kicks off tomorrow 🎉
📍 The Midway, SF, 6pm 👇
na2.hubs.ly/H08nMYB0
Join us alongside other key players in AI helping organizations adopt and scale their use of open-source.
ray retweeted
We post-train GLM-5.3 Flash in a sandbox that reproduces the Bits Investigation tools over archived scenario data.
Our async RL stack is based on #SkyRL, with efficiency and reliability enhancements in our internal fork. Training and inference run concurrently, with a staleness budget for off-policy updates.
ray retweeted
datadog ai research used SkyRL to post-train GLM 5.3 Flash to perform at the pareto frontier for their SRE eval
@bthecohen and team were some of the earliest users of SkyRL at scale, really cool to see these results finally out! great tech report as well
Meet Scout 1.0, our first AI model built for investigating production incidents.
Scout matches frontier-model investigation quality at a fraction of the cost, which makes it practical to investigate more issues, earlier.
We trained Scout on realistic simulations built from real-world incidents and production telemetry, using reinforcement learning to reward accurate root causes backed by evidence.
Learn more and sign up for the preview: bit.ly/3TJknGO
ray retweeted
Great to see this out!
Datadog's new SOTA model for investigating production incidents. Their in-house training stack builds on top of SkyRL! 🚀🚀🚀
Meet Scout 1.0, our first AI model built for investigating production incidents.
Scout matches frontier-model investigation quality at a fraction of the cost, which makes it practical to investigate more issues, earlier.
We trained Scout on realistic simulations built from real-world incidents and production telemetry, using reinforcement learning to reward accurate root causes backed by evidence.
Learn more and sign up for the preview: bit.ly/3TJknGO
ray retweeted
Datadog post-trained GLM-5.3 Flash via RL on real-incident sims. Matches frontier SRE performance at much lower cost.
Their RL stack is built on top of SkyRL! Check out the technical report web-assets.dd-static.net/425…
Meet Scout 1.0, our first AI model built for investigating production incidents.
Scout matches frontier-model investigation quality at a fraction of the cost, which makes it practical to investigate more issues, earlier.
We trained Scout on realistic simulations built from real-world incidents and production telemetry, using reinforcement learning to reward accurate root causes backed by evidence.
Learn more and sign up for the preview: bit.ly/3TJknGO
ray retweeted
@ChefRobotics trains the Vision-Language-Action models behind its food assembly robots 10x faster on Anyscale, with 92% scaling efficiency across multi GPU runs on 16 A100s.
Read the story: na2.hubs.ly/H08q5bh0
ray retweeted
@Coinbase is building an agent-first ML platform where AI agents watch model performance and trigger retraining on their own.
Aman Choudhary on how Ray on Anyscale helps them run up to 1 billion inferences a day
na2.hubs.ly/H08nMPc0
ray retweeted
We’re joining Open Together with @huggingface and friends this Friday to support everyone building AI with open-source
SF, Oct 16, 6pm 👇
na2.hubs.ly/H08m4Mv0
ray retweeted
Our upcoming webinar will cover how to design scalable distributed inference systems for production AI applications, from classic ML models to large language models.
Learn how Ray Serve delivers high-throughput, low-latency serving, and how to evaluate the performance tradeoffs in autoscaling, batching, routing, and inference-engine optimization.
Topics include:
- Building Ray Serve applications with deployments, composition, and load-aware request routing
- Deploying and scaling LLM endpoints with Ray Serve LLM on engines like @vllm_project and @sgl_project
- Optimizing the inference engine with the KV cache, paged attention, prefix caching, and continuous batching
- Scaling frontier models with KV-cache-aware routing, prefill/decode disaggregation, and wide expert parallelism for mixture-of-experts (MoE) models
- Deploying Custom LLMs on Anyscale and Integrating Them with Agentic Platforms (Claude Code, Cursor, Codex)
- Comparing costs across self-hosted GPUs, subscriptions, and token-based API billing.
Register: na2.hubs.ly/H08hfbl0
ray retweeted
We’re heading to Long Horizon by Encord next week!
📍Come meet our team at booth P1 and hear from our team:
🎤 From Petabytes to Policies: Scaling the Physical AI Data Engine with Ray
🎤 Why Sim Platforms Shouldn’t Be a Batch Job: Scaling Synthetic Data Generation with Ray
na2.hubs.ly/H08jzKS0
ray retweeted
Anyscale is heading to PyTorch Conference North America, October 20 and 21 in San Jose. Catch our sessions on SkyRL and Ray Core, drop by our Meet the Developers session on DeepSpeed and Ray with the Anyscale team, swing by booth P9, and come recharge with us at the Coffee Club lounge just off the venue.
na2.hubs.ly/H08nkrd0
ray retweeted
People who are really serious about applied AI should train their own models.
Sharing our “Owning Intelligence” series on how we built, evaluated, and post-trained a fraud agent at Coinbase—from using frontier models to owning a model optimized for our task.
Part 1 — Build
lnkd.in/grtg_gfF
Part 2 — Evaluate
lnkd.in/g-6DKdZ8
Part 3 — Own
lnkd.in/g9ET_4at
@akshit_trehan
@anyscalecompute, @erictang000 , @sumanthrh , @seiji_________ ,
ray retweeted
@Xoople is building toward petabyte-scale processing of satellite imagery of the whole planet. VP of Engineering Milos Colic on why the team runs Ray on Anyscale instead of building its own AI compute layer.
na2.hubs.ly/H08gWwX0
ray retweeted
After a year of close collaboration with the Azure team, we're excited to announce the general availability of Anyscale on Azure! Anyscale on Azure is managed Ray on AKS, for the whole loop from data processing to training to inference. Some highlights:
- Deploys on AKS in your own Azure subscription
- Data and models stay in your subscription
- A first-class resource in the Azure portal
- Sign in with Microsoft Entra ID
- Billed through your Azure invoice, counting toward your existing Azure consumption commitment
This is the product of a very deep collaboration with the AKS team and is just the start of what we'll be shipping on Azure.
Learn more: anyscale.com/product/azure
New: Anyscale on @Azure, now in GA.
Anyscale on Azure is generally available. Run the full AI loop on AKS in your own subscription. Own your AI on open source Ray, keep data and models in your tenant, and get more work from every GPU. Deploy from the Azure portal and pay through your Azure bill.
na2.hubs.ly/H08dWKQ0
ray retweeted
New: Anyscale on @Azure, now in GA.
Anyscale on Azure is generally available. Run the full AI loop on AKS in your own subscription. Own your AI on open source Ray, keep data and models in your tenant, and get more work from every GPU. Deploy from the Azure portal and pay through your Azure bill.
na2.hubs.ly/H08dWKQ0