CentML provides a platform for deploying Large Language Models (LLMs) quickly, offering performance preview, resource optimization, and automated LLM optimizations.

CentML

We offer an optimization platform for AI deployment. Using our platform, you significantly save on your costs for small and gigantic models.

Radical Ventures

Ivan Zhang

Pavel Stoev

Deloitte Ventures

Thomson Reuters Ventures

Gradient Ventures

AAL VC

NVIDIA

Misha Bilenko

It's been a busy summer at CentML ☀ 🏖 ! We're excited to welcome FIVE amazing new additions to our team. 

Jeannette B., Office Manager - Palo Alto
Zhumakhan Nazir, Software Development Engineer
Ali Jahanshahi, Software Development Engineer
Vasilis Vagias, Staff Solutions Engineer
T. Wood, Content Manager

Welcome, everyone! 🎉 

Check out CentML's open roles at https://centml.ai/careers/

Release of AI at Meta's Llama-3.1-405B today!  

We are excited to announce CentML's recent contributions to vLLM, which adds pipeline parallel inference support. This addition has made significant performance improvements when deploying Llama-405B on multi-node with vLLM. 

We are committed to collaborating with the open-source community to make ML more accessible, affordable, and efficient, both for open-source solutions like vLLM, as well as enterprise ones using CentML's CServe.

Read more: 
https://lnkd.in/g3qrQh4i

From Mexico to India and nearly everywhere in between, people across the globe are running Llama 3.1 using CentML 🌍🦙

This marks another step forward in our mission to democratize AI and make leveraging LLMs more sustainable. Exciting times ahead!

Get your hands on Llama 3.1-405B using CServe here: https://lnkd.in/gCkj7qPZ

With CServe, you’ll receive unparalleled model integration:

💰 Serving costs reduced by up to 65%
🏎️ Lightning-fast AI capabilities
🦾 Cutting-edge AI integration
⚡️ Single-click deployment

Are you a current CentML user? You can make the switch from previous Llama versions with a single line of code.

#machinelearning #opensourceAI #llama3 #CentML #CServe

Are you overpaying for GPUs?

Discover how DeepView can help you:
- Predict model performance across various cloud GPUs
- Make informed decisions between cost-effective options and high-end GPUs like H100
- Identify and resolve performance bottlenecks

Don't let the allure of the latest GPUs drain your budget. Learn how to optimize your ML workflows and potentially save thousands in cloud computing costs.

Read the full post: [https://lnkd.in/gFEtSJjY]

#MachineLearning #CloudComputing #GPUOptimization #AI #DeepLearning

Great job of the CentML team to make it easier for the vLLM community to run large LLama-405B efficiently! #centml #vllm #genai #optimization

Check out the tech deep dive into the pipeline parallel inference implementation which #CentML contributed to #vLLM! The code that we wrote are currently used by Snowflake & others who deploy Llama 3.1-405B through vLLM.

Special shout-out to the one-and-only Kaichao You who spent so much of his time supporting this work :)

AI:CENTML

CentML