- April 14, 2026
- 3 min read
FriendliAI and Samsung Cloud Platform Forge Strategic Alliance to Power Frontier Model AI Inference on NVIDIA B300 GPUs
- FriendliAI is collaborating with Samsung SDS, a leading GPU infrastructure-as-a-service (IaaS) provider in South Korea, to deliver frontier model AI inference services to global startups and enterprises.
- The initiative brings together FriendliAI's high-performance inference stack and Samsung Cloud Platform (SCP)'s scalable NVIDIA B300 GPU infrastructure.
FriendliAI is collaborating with Samsung SDS, a leading GPU infrastructure-as-a-service (IaaS) provider in South Korea, to deliver frontier model AI inference services to global startups and enterprises. The initiative brings together FriendliAI's high-performance inference stack and Samsung Cloud Platform (SCP)'s scalable NVIDIA B300 GPU infrastructure.
FriendliAI is known for building an AI inference platform that delivers unmatched speed, cost efficiency, and reliability. By integrating FriendliAI’s platform with Samsung SCP’s B300 GPU infrastructure, the two companies will enable customers to run the latest frontier open-weight models — including GLM-5.1, MiniMax M2.5, NVIDIA Nemotron 3 Super, and DeepSeek v3.2 at maximum performance with production-grade reliability and competitive token pricing.
Bringing Frontier Open-Weight Model AI Inference to Global Companies
The partnership is designed to provide organizations worldwide with a seamless and scalable experience when deploying frontier AI models. Customers will be able to access FriendliAI’s inference platform powered by Samsung SCP’s NVIDIA B300 GPU IaaS, combining state-of-the-art inference optimization with high-performance GPU infrastructure. Key benefits include:
Day-0 Support for Frontier Open-Weight Models: Immediate support for models such as GLM-5.1, MiniMax M2.5, NVIDIA Nemotron 3 Super, and DeepSeek v3.2 enables companies to stay at the forefront of AI innovation without building and maintaining custom inference stacks.
High-Performance Inference: FriendliAI’s optimized platform is built on its in-house inference engine featuring custom kernels, advanced quantization techniques, intelligent caching, and adaptive speculative decoding.
Cost-Effective Scalability: The partnership enables high-speed inference with low operational cost and maximum scalability. Friendli Serverless Endpoints provide a flexible, token-based pricing model and pay only for the tokens they process.
Global Reach and Reliability: FriendliAI’s inference infrastructure, integrated with Samsung SCP’s B300 IaaS platform, provides a global footprint with high-availability guarantees, ensuring low-latency and stable service delivery for customers worldwide.
“FriendliAI is thrilled to partner with Samsung SCP to empower AI inference for companies around the world,” said Byung-Gon Chun, Founder and CEO of FriendliAI. “Our inference optimization technology, built for unmatched speed and cost efficiency, perfectly complements Samsung SCP’s cutting-edge NVIDIA B300 GPU cloud infrastructure. This collaboration allows us to deliver the latest frontier open-weight models at full performance, helping customers unlock new business opportunities with agentic AI.”
EUNYOUNG KIM, Executive Vice President at Samsung SDS, added, “Our mission at Samsung SCP is to provide flexible and high-performance GPU infrastructure for the AI era. Partnering with FriendliAI enables us to bring state-of-the-art AI inference capabilities to global customers. FriendliAI’s platform ensures that our NVIDIA B300 GPUs are fully utilized, offering clients the optimal combination of performance, cost efficiency, and reliability for frontier AI workloads.”
About FriendliAI
FriendliAI is The Frontier AI Inference Cloud. Built by the researchers who invented continuous batching, FriendliAI’s highly optimized inference engine efficiently runs state-of-the-art open-weight and custom models at production scale with 99.99% reliability. By maximizing GPU utilization, FriendliAI delivers speeds up to 3x faster than vLLM and 50% to 90% cost savings relative to closed model APIs, empowering engineers to deploy frontier AI with uncompromising speed and model ownership. Learn more at https://friendli.ai
About Samsung Cloud Platform (SCP)
Samsung Cloud Platform is a leading GPU infrastructure-as-a-service (IaaS) operating clusters based on NVIDIA B300 GPUs. Samsung SCP offers scalable, high-availability GPU resources designed to accelerate compute-intensive workloads such as AI inference and training. Its mission is to empower organizations with flexible, cost-effective infrastructure that keeps pace with the rapid evolution of artificial intelligence. Learn more at https://cloud.samsungsds.com/
For more information about the FriendliAI and Samsung SCP partnership or to inquire about services, please contact:
FriendliAI: press@friendli.ai
Samsung Cloud Platform: press@scp.com
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.
