- November 19, 2025
- 2 min read
FriendliAI Partners with Nebius to Deliver High-Performance, Cost-Efficient AI Inference
We are thrilled to announce a strategic partnership with Nebius, a leading global GPU cloud Infrastructure provider. This collaboration is a significant step in our mission to make world-class, high-performance AI inference technology universally accessible and affordable for AI startups and Enterprises worldwide.
FriendliAI’s proprietary inference optimization technology has been integrated into Nebius' large-scale AI cloud infrastructure. This powerful combination is set to deliver an immediate upgrade to companies running essential generative AI services including customer support chatbots, coding copilots, and AI agents on the FriendliAI platform.
Why This Partnership Matters: The Ultimate AI Efficiency Stack
For AI startups and enterprises, the cost and operational complexity of running AI models at scale are major burdens. Our partnership directly tackles these challenges by integrating our unique AI inference platform with Nebius' robust GPU infrastructure, providing unmatched performance across three critical metrics:
- Cost Efficiency: Customers can achieve over 50% reduction in GPU costs, a substantial saving for high-volume AI workloads.
- Blazing Speed: Leveraging our purpose-built inference stack, we provide over 2x faster inference speed.
- Unwavering Stability: Our combined solution is engineered for reliability, offering a 99.99% uptime guarantee (SLA) that is crucial for mission-critical enterprise applications.
- Guaranteed GPUs for Instant Scaling: Expanding our multi-cloud strategy by adding Nebius' reliable infrastructure, FriendliAI automatically scales GPU resources to match fluctuating inference demand, optimizing performance and cost. We also guarantee GPU availability for sudden spikes or large workloads, preventing the throttling and service interruptions.
By utilizing FriendliAI’s API on the Nebius infrastructure, AI startups and enterprises can instantly experience improved speed, superior cost efficiency, and enhanced stability in their inference environments.
Meet Nebius: A Global Leader in GPU Infrastructure
Nebius is a neo-cloud company based in Amsterdam, providing infrastructure tailored for the AI workloads. As a publicly-traded company on NASDAQ, Nebius currently provides high-performance AI workload infrastructure across Europe, North America, and Israel.
Notably, Nebius recently solidified its position as a core supplier in the global AI infrastructure market by entering into a significant $19.4 billion AI computing partnership with Microsoft.
This partnership ensures that customers deploying AI models will benefit from an ecosystem built on global-scale reliability and top-tier performance.
Quote from FriendliAI CEO
"Our goal is to make our world-class AI inference technology easily accessible to all companies," said Byung-Gon Chun, CEO of FriendliAI. "The integration of FriendliAI's inference optimization technology with Nebius' robust GPU cloud means that every customer can now deploy AI models with a combination of top-level latency, stability, and cost-efficiency."
We are dedicated to delivering the best-in-class inference platform that powers real-world, production-ready AI applications. The partnership with Nebius accelerates this vision, making high-performance AI inference a global reality.
Written by
FriendliAI Tech & Research
Share
General FAQ
What is FriendliAI?
FriendliAI is the Frontier Inference Cloud for Agents, delivering high throughput, low latency, and reliability at scale for agentic workloads. Through vertically optimized inference infrastructure, it delivers 2–5× faster output token speed and a 99.99% uptime SLA for high-volume production traffic.
How does FriendliAI reduce inference costs?
FriendliAI reduces inference costs through higher GPU utilization and optimized inference performance. FriendliAI's patented continuous batching technique, along with quantization, speculative decoding, KV cache offloading, multi-LoRA serving, and autoscaling, helps you serve more tokens with fewer GPUs, lowering your infrastructure costs without sacrificing performance.
Why should I choose FriendliAI over other inference providers?
FriendliAI is built for production AI agents, combining speed, reliability, and efficiency at scale. It delivers low-latency streaming, reliable long-context inference, and robust tool calling without compromising stability. According to independent OpenRouter benchmarks, FriendliAI consistently ranks among the top providers for throughput, latency, and reliability across leading open-weight models. See why customers choose FriendliAI
Which open-weight models does FriendliAI support?
Run today’s frontier open-weight models—including GLM, MiniMax, Kimi, DeepSeek, Qwen, Gemma, and more—with a simple API call. FriendliAI Model API gives you instant access to the latest models with optimized inference performance for production workloads. Explore models and pricing
How do I get started?
Getting started takes just a few minutes. [1] Sign up for FriendliAI, [2] Generate your API key, and [3] Make your first inference request with frontier open-weight models.
Still have questions?
If you want a customized solution for that key issue that is slowing your growth, support@friendli.ai or click Talk to an engineer — our engineers (not a bot) will reply within one business day.
