Ori

📢 Announcing Ori Inference Endpoints, an easy and scalable way to deploy state-of-the-art machine learning models as API endpoints. ⚡ Deploy your favorite AI model in a single click. Ori Inference Endpoints makes inference effortless. 📊 Automatically scale inference up or down based on demand, from thousands of GPUs all the way down to zero. ⚖️ Per-minute pricing helps you keep your inference infrastructure affordable and costs predictable. Discover more about Ori Inference Endpoints 🔗 https://lnkd.in/gvjmNDU8

✨ Happy Holidays from all of us at Ori!🌟 As 2024 winds down, we want to take a moment to thank our valued customers, partners, and the entire Ori community for an extraordinary year of growth and collaboration. Wishing you a joyful holiday season and a Happy New Year! 🎉

We're excited to announce that Jacob Smith has joined Ori's board of directors! Jacob brings a wealth of experience in scaling tech companies and fostering innovation in the digital infrastructure space. Jacob's insights will be instrumental as we continue to empower AI businesses with cutting-edge infrastructure. Welcome to the Ori family, Jacob! https://lnkd.in/dEuaX-p5

We are thrilled to share that Ori will be the first AI infrastructure provider to deploy NVIDIA's H200 GPUs in a UK data centre. This milestone reflects our commitment to continue providing cutting-edge solutions that empower businesses to achieve their AI ambitions. The H200, with 141GB of HBM3e memory, offers almost twice the capacity of the H100, while its 4.8TB/s bandwidth delivers 1.4x greater performance. If you'd like to try them out, contact us! https://lnkd.in/ek5Q2YAg

Llama 3.3 70B, AI at Meta's newest model is as powerful as the flagship #Llama 3.1 405B, providing state-of-the-art performance with a nimble footprint. Looking to try Llama 3.3 on a cloud GPU instance? Here's how you can do it in just a few short steps - https://lnkd.in/g_y9XdSV #llm #GenAI

Ori's CEO, Mahdi Yahya, makes a compelling case for establishing Special Compute Zones to help the UK develop sovereign cloud infrastructure and harness its unique strengths in AI. In his latest piece for DatacenterDynamics, he explains why modernizing power and cloud infrastructure is vital for the UK to realize its ambitious AI goals and stay competitive on the global stage. Read the full article here: https://lnkd.in/gsP826mC

Hi all! At Ori, we are hiring our first Graphic Designer to join our amazing marketing team at Ori. You will work together with Daniel Van den Berghe to manage and execute the design function for Ori. This a dynamic, hands-on role where no two days are the same! We are growing and have a lot of exciting things coming up in 2025! 🌟 Click the job description in the comments for more info and details on how to apply 📩

Love the simplicity of Ollama but need to scale automatically for larger workloads? With KEDA (Kubernetes Event-Driven Autoscaling) and Ori Serverless Kubernetes, you can seamlessly scale your LLM deployments based on GPU usage—enabling optimal performance without having to manage infrastructure yourself. 🔗 Explore our step-by-step tutorial to see how easy scaling can be: https://lnkd.in/gxXixZbM

As GPUs get more powerful, it's equally important to have ample and fast memory to make the most of performance gains. That’s why we’re excited about NVIDIA’s H200 GPU. With 76% more memory and 43% higher bandwidth compared to the H100, it's built to handle today’s larger models and deliver faster inference. Curious how the H200 stacks up? Check out our blog for all the details, from features to use cases and comparisons: https://lnkd.in/degWPdbB #nvidia #h200 #gpu

In this article for Verdict, Ori CEO Mahdi Yahya underscores the critical need for UK's technology leaders to champion British cloud providers. By prioritizing homegrown cloud infrastructure, the UK can drive innovation, ensure data sovereignty, and significantly reduce cloud expenditure. Read the full article 🔗 https://lnkd.in/gesYm-qv