Lablup to Showcase Backend.AI’s GPU Virtualization and Edge Scalability at ISC 2025

Cloud Native Computing Foundation Adds 46 New Members

Accelerating Hyperparameter Tuning with Container-Level GPU Virtualization

Backend.AI in the News

**\- Operated a 60+ node B200 cluster for 73 days, cutting average recovery time by 47%**

**\- Shares real-world operational strategies from the Solar Open 100B large-model training project**

**\- Live demos of Backend.AI Continuum and Backend.AI:GO on NVIDIA DGX Spark systems**

SEOUL, South Korea – March 16, 2026 — Lablup Inc., a leading AI infrastructure company, will share its experience operating a large-scale Sovereign AI training environment powered by 504 NVIDIA B200 GPUs at NVIDIA GTC 2026, held March 16–19 in San Jose, California. During the GTC Theater Session on March 18, CEO Jeongkyu Shin will present insights from 73 days of large-cluster operations, covering Lablup’s fault-tolerant scheduling strategies and fast recovery techniques. The company will also host hands-on demonstrations at its booth, featuring automatic recovery functions in Backend.AI Continuum and Backend.AI:GO running on NVIDIA DGX Spark systems.

**Sovereign AI in Action: Building a 100B-Parameter Model**

In the session titled “Building Sovereign AI: Scaling 100B+ Model Training on NVIDIA Blackwell Infrastructure,” Shin will walk through how Lablup trained a 100-billion-parameter model on a 60+ node B200 cluster (504 GPUs). This project was part of the Korea Ministry of Science and ICT / NIPA initiative to develop homegrown AI foundation models, with Lablup serving as the infrastructure partner for Upstage’s Solar Open 100B training.

The team built a fault-tolerant scheduling system capable of automatically detecting and recovering from common distributed training failures such as GPU errors and NCCL timeouts. The system reduced the average recovery time by 47% compared to previous runs, restarting failed processes in under three seconds. Lablup will also share its experience troubleshooting issues like a single NFS driver misconfiguration that degraded performance by nearly tenfold, and its optimizations for MXFP8 training stability and NCCL tuning across both RoCE and InfiniBand networks adapted for the Blackwell architecture.

**Hands-On Demos: Backend.AI Continuum and Backend.AI:GO**

Visitors to Lablup’s booth (#243) can take part in live demonstrations showcasing Backend.AI Continuum’s model routing and resilience features. For example, when a network cable is unplugged to simulate a fault, inference requests seamlessly reroute through alternate paths in real time. Even if a cloud connection drops, Continuum maintains API calls by switching instantly to local resources, allowing attendees to physically experience its fault tolerance.

Lablup will also present Backend.AI:GO, a local AI runtime that runs on devices ranging from personal laptops to high-end systems like the NVIDIA DGX Spark with 128GB unified memory. Backend.AI:GO provides a personal, high-performance AI environment that fully utilizes local hardware without depending on remote cloud access.

“Running a 504-GPU B200 cluster continuously for 73 days taught us what truly breaks in large-scale distributed training and how to rebuild it,” said Lablup CEO Jeongkyu Shin. “At GTC 2026, we’re excited to share those lessons and our vision for building sovereign AI infrastructure that enables nations and industries to operate AI on their own terms.”

**About Lablup Inc.**

Lablup Inc., founded in 2015, builds Backend.AI, a software-defined AI infrastructure platform that orchestrates heterogeneous GPU and NPU clusters at hyperscale with secure multi-tenancy and multi-node workload management. Its container-level GPU virtualization technology further maximizes accelerator utilization across training, inference, and deployment. Headquartered in Seoul with a U.S. subsidiary in San Jose, Lablup manages over 16,000 GPUs across 110+ sites worldwide, and Backend.AI is validated as NVIDIA DGX-Ready Software. Learn more at lablup.com.

Lablup to Showcase Sovereign AI Training Infrastructure at NVIDIA GTC 2026 Powered by 500+ B200 GPUs

*   Lablup and UK-based HPC specialist Boston Limited enter a strategic distribution partnership to bring the Backend.AI platform to the EMEA market.
*   Lablup Partners with Boston Limited to Distribute Backend.AI Across EMEA

**SEOUL, SOUTH KOREA & MWC 2026 | BARCELONA, SPAIN – March 2, 2026** – Lablup Inc., the AI infrastructure operating system company, and Boston Limited, the UK-based high-performance computing specialist, today announced a strategic partnership to distribute and sell the Backend.AI platform across the EMEA region.

Under the agreement, Boston Limited will offer Backend.AI cloud services (SaaS) and on-premises software licenses directly to end customers in EMEA, along with installation, configuration, and consulting services. The two companies plan to build local sales and technical support operations through Boston Limited's established customer network and HPC infrastructure distribution capabilities across the region.

Backend.AI is a software-defined AI infrastructure platform that manages heterogeneous accelerators, including chips from NVIDIA, AMD, Intel, and domestic NPU vendors, through a single control plane. Organizations can operate an entire cluster of mixed GPU and NPU hardware as one unified resource pool. Container-level GPU virtualization and the Sokovan orchestrator, both developed in-house, give EMEA customers:

*   A single platform to manage heterogeneous GPU and NPU hardware from multiple vendors
*   Local installation, configuration, and consulting support through Boston Limited's established regional network
*   Integrated operation across existing HPC infrastructure, reducing the cost and operational complexity of deploying AI at enterprise scale

Boston Limited has supplied HPC infrastructure throughout the UK and internationally for over 30 years. The company delivers custom high-performance computing solutions (servers, storage, GPU clusters) to research institutions, universities, enterprises, and public-sector organizations, and maintains close partnerships with Supermicro, NVIDIA, Intel, and AMD.

Lablup established a U.S. subsidiary in San Jose in 2025 to enter the North American market. With this new partnership, Backend.AI's commercial distribution now spans Asia, North America, and the EMEA region.

"Expanding Backend.AI's reach into EMEA required a partner with deep knowledge of the local market," said Jeongmook Kim, CRO at Lablup Inc. "Boston Limited's three decades of HPC infrastructure distribution and broad customer network across the region made them the right fit. Lablup is now working closely with Boston Limited to deliver strong technical support for customers on the ground."

"As AI workloads grow, our customers increasingly need software platforms that let them manage GPU resources efficiently," said Manoj Nayee, Managing Director at Boston Limited. "Adding Backend.AI to the HPC and AI infrastructure we already supply gives EMEA customers a complete solution from hardware to software management. For organizations running GPUs from multiple vendors, this integration can meaningfully improve the return on their infrastructure investment."

![MWC_2026_BOSTON_PARTNERSHIP_VERTICAL.jpg](https://cdn.lablup.com/MWC_2026_BOSTON_PARTNERSHIP_VERTICAL_7377c1942c.jpg)

The partnership was signed on March 2 at MWC 2026, held at Fira Gran Via in Barcelona, Spain.

---

### About Lablup Inc.

Lablup Inc., founded in 2015, builds Backend.AI, a software-defined AI infrastructure platform that orchestrates heterogeneous GPU and NPU clusters at hyperscale with secure multi-tenancy and multi-node workload management. Its container-level GPU virtualization technology further maximizes accelerator utilization across training, inference, and deployment. Headquartered in Seoul with a U.S. subsidiary in San Jose, Lablup manages over 16,000 GPUs across 110+ sites worldwide, and Backend.AI is validated as NVIDIA DGX-Ready Software. Learn more at lablup.com.

### About Boston Limited

Trusted for over 30 years and operating in more than 40 countries, Boston Limited is a global provider of co-designed, custom and certified IT systems powering next-generation intelligent infrastructure. Its portfolio spans edge devices, desktops, servers, storage, clusters and reference architectures, delivered through an engineering-led, customer-centric engagement model and available via boston.co.uk.

Over three decades, Boston has continually evolved to meet the changing demands of the technology landscape - from component distribution to value-added system integration, and today as a leader in data, high-performance computing (HPC) and AI solutions. Boston Limited is part of US-based SourceCode, LLC.

 Lablup Partners with Boston Limited for EMEA Distribution of Backend.AI

_**Backend.AI integrates with VAST AI Operating System to deliver high-performance data access for AI workload orchestration across training, deployment, and inference**_

**SEOUL, SOUTH KOREA & VAST Forward | SALT LAKE CITY, UT – February 25, 2026** – Lablup Inc., the AI infrastructure operating system company, today announced it has joined VAST Constellation as a Technology Alliance Partner, collaborating with VAST Data, the AI Operating System company, to help customers accelerate AI training and inference workloads through a validated integration between Backend.AI and the VAST AI Operating System.

As enterprises race to operationalize AI, proven partner integrations have become one of the fastest ways to deliver validated architectures, reduce deployment risk, and accelerate time-to-value. Through Constellation, VAST and Lablup are able to build, validate, deliver, and scale AI-ready solutions across the data center, cloud, and at the edge.

"Lablup is committed to democratizing AI infrastructure through accessible, efficient, and scalable technology," said Jeongkyu Shin, CEO at Lablup Inc. "By integrating Backend.AI with the VAST AI OS, developers and enterprises can now handle large-scale datasets and complex workloads while maintaining the reliability and high performance that modern AI demands. Storage infrastructure matters just as much as the orchestrator managing AI workloads, and Lablup is dedicated to delivering a unified environment that supports both seamlessly."

With this collaboration, Lablup integrates Backend.AI with the VAST AI OS to enable customers to orchestrate end-to-end AI workloads, from training to deployment and inference, with high-throughput access to training datasets and model artifacts through RDMA and GPU-direct data paths powered by VAST DataStore. This integration streamlines data access, reduces latency, and maximizes GPU utilization across heterogeneous compute environments.

By delivering a validated solution combining Backend.AI's GPU orchestration and workload management capabilities with VAST's Disaggregated Shared-Everything (DASE) architecture and unified data services, organizations can achieve:

\- Faster deployment and simplified integration for AI pipelines through validated architectures that reduce implementation complexity

\- Improved performance and predictable outcomes under sustained workloads with high-throughput data access and GPU-direct connectivity

\- Consolidated infrastructure management that reduces operational overhead while scaling AI training, inference, and deployment environments

"Enterprises are looking for a simpler, more consistent way to deploy production AI, backed by validated integrations they can trust," said John Mao, Vice President, Global Technology Alliances at VAST Data. "We're proud to welcome Lablup into VAST Constellation, expanding the ecosystem of industry-leading solutions built on the VAST AI Operating System and helping customers reduce integration risk while accelerating AI-ready deployments from the data path to the application layer."

Lablup and VAST Data are collaborating on infrastructure for South Korea's Sovereign AI foundation model initiatives, and this Constellation partnership further strengthens their joint commitment to delivering production-ready AI infrastructure solutions.

**About Lablup**

Lablup Inc., founded in 2015, builds Backend.AI, a software-defined AI infrastructure platform that orchestrates heterogeneous GPU and NPU clusters at hyperscale with secure multi-tenancy and multi-node workload management. Its container-level GPU virtualization technology further maximizes accelerator utilization across training, inference, and deployment. Headquartered in Seoul with a U.S. subsidiary in San Jose, Lablup manages over 16,000 GPUs across 110+ sites worldwide, and Backend.AI is validated as NVIDIA DGX-Ready Software. Learn more at lablup.com.

Lablup Joins VAST Constellation to Accelerate AI Training and Inference with Integrated Infrastructure

**The free desktop application runs small language models entirely offline on personal PCs and supports multi-device distributed inference through Mesh Mode. Lablup will also sign a partnership with UK-based AI infrastructure firm Boston Limited during the event.**

**SEOUL —** Lablup Inc. today announced it will showcase Backend.AI:GO, its free personal AI platform, at MWC Barcelona 2026, held March 2 through 5 at Fira Gran Via in Barcelona, Spain. The company will also hold a partnership signing ceremony with Boston Limited, a UK-based AI infrastructure company, as it formally enters the European market.

Backend.AI:GO, launched at CES 2026 in January, is a free desktop application that runs small language models (SLMs) directly on a personal PC with no cloud connection required. Once a model is downloaded, the application operates entirely offline, handling document analysis, image understanding, and code review without transmitting data externally. It supports macOS, Windows, and Linux, with native Apple Silicon optimization through Lablup's own MLxcel inference engine.

When a task exceeds local hardware capacity, Backend.AI:GO can automatically offload it to a Backend.AI Core cluster within the organization or to cloud APIs from providers such as OpenAI and Google. Users can also link multiple GO devices together through Mesh Mode, pooling computing resources for distributed inference on models too large for a single workstation.

Since launch, Backend.AI:GO has added capabilities at a rapid pace. Version 1.1, released in early February, introduced Stable Diffusion-based image generation, document translation with OCR support for TXT, MD, DOCX, PDF, and image files, and a headless server mode for standalone deployment in server environments. Version 1.2 adds an agent marketplace where users can browse and download agent profiles, along with a Cowork mode that lets multiple agents collaborate under shared folder permissions and automated approval rules.

**From Desktop to Mobile**

Lablup is developing the next phase of Backend.AI:GO: the ability to remotely access and control AI models running on a desktop, laptop, server, or NVIDIA DGX Spark from a mobile device. Any machine running GO becomes a personal AI server that a smartphone can connect to for conversations or task requests. The orchestration capabilities that Lablup has built for enterprise AI infrastructure now extend to individual users and their personal devices.

"Backend.AI:GO was the first step in bringing AI workloads that once required a data center onto personal devices, and now Lablup is extending that reach to mobile," said Jeongkyu Shin, CEO of Lablup Inc. "There is no reason anyone should be tied to a specific location or a specific machine to use AI. That has been the direction Backend.AI has pursued from the beginning."

Lablup will be exhibiting at MWC Barcelona 2026 in Hall 7, Booth #7A62-01.

**About Lablup Inc.**

Lablup Inc., founded in 2015, builds Backend.AI, a software-defined AI infrastructure platform that orchestrates heterogeneous GPU and NPU clusters at hyperscale with secure multi-tenancy and multi-node workload management. Its container-level GPU virtualization technology further maximizes accelerator utilization across training, inference, and deployment. Headquartered in Seoul with a U.S. subsidiary in San Jose, Lablup manages over 16,000 GPUs across 110+ sites worldwide, and Backend.AI is validated as NVIDIA DGX-Ready Software. Learn more at lablup.com.

Lablup Brings Personal AI Platform Backend.AI:GO to Europe at MWC Barcelona 2026

_\- Enterprise AI infrastructure now extends to personal PCs while maintaining centralized management_  
_\- Cloud version Backend.AI:DOL also launches, completing full-stack coverage from AI PCs to hyperscale clusters_

![(Lablup) CES 2026.jpg](https://cdn.lablup.com/Lablup_CES_2026_491d15e4a7.jpg)

**LAS VEGAS, January 6, 2026** – Lablup, an AI infrastructure platform company, today announced the official launch of Backend.AI:GO at CES 2026. The desktop application enables users to run Small Language Models (SLMs) directly on personal computers without cloud servers. This marks Lablup's fourth consecutive year exhibiting at CES.

Backend.AI:GO (where "GO" stands for "Generative On-device") accelerates AI inference using local computing resources in on-premises environments. Once users download a model, the application operates entirely offline without requiring an internet connection. User data is never transmitted to external servers. Users can securely download models to their personal computers and freely utilize them for internal document analysis, image understanding and generation, code review, and more. The application also allows users to adjust hyperparameters and resource allocation to customize model response characteristics and resource usage.

The key differentiator of Backend.AI:GO lies in its integration with Lablup's AI infrastructure management platform, Backend.AI. By connecting multiple Backend.AI:GO instances or integrating with existing Backend.AI environments and cloud AI services such as OpenAI and Google, users can leverage both large-scale AI models and local PC resources in a unified manner. Organizations can extend the same management approach used for thousands of GPU clusters to AI PCs, enabling centralized model deployment, version control, and usage monitoring from a single console.

Lablup will also officially launch Backend.AI:DOL at CES 2026. Previously introduced as a beta version at AI Infra Summit, Backend.AI:DOL (short for "Deployable Omnimedia Lab") is a cloud-based solution that allows users to access various open models directly through a web browser. With Backend.AI:GO handling on-device workloads and Backend.AI:DOL managing cloud deployments, Lablup now offers a complete full-stack AI infrastructure solution spanning from AI PCs to hyperscale GPU clusters.  
"As the AI PC era accelerates, enterprise demand for running LLMs directly in on-device environments is growing rapidly," said Jungkyu Shin, CEO of Lablup. "With Backend.AI:GO and DOL, we will provide an environment where organizations can operate AI infrastructure with a consistent experience, regardless of scale, from a single PC to thousands of GPU clusters."

Shin added, "Lablup has expanded its presence in the global AI infrastructure market through four consecutive years of CES participation. At this year's CES, we are introducing a full-stack AI infrastructure solution that spans on-device and cloud, setting a new standard for AI infrastructure operations."

Lablup is exhibiting at CES 2026 in LVCC North Hall, Booth #9529, from January 6–9, 2026.

**About Lablup**  
Lablup Inc. is redefining how modern AI systems are powered by delivering the operating model for scalable, high-performance GPU computing. Its flagship platform, Backend.AI, combines software-defined GPU virtualization, ultra-fast containerized environments, and intelligent workload orchestration to maximize performance, utilization, and efficiency across heterogeneous clusters.

Founded in 2015, Lablup works with global technology leaders, research institutions, and top universities to advance AI infrastructure and accelerate data-driven innovation worldwide. Learn more at lablup.com and follow Lablup on LinkedIn.

Lablup unveils 'Backend.AI:GO' at CES 2026: Run LLMs on personal computers without cloud connectivity

**SAN FRANCISCO, CA** – **November 14, 2025** – Lablup Inc., a leader in AI infrastructure solutions, continues to revolutionize the AI education and research landscape through its Backend.AI platform. At SC25 in St. Louis, Missouri, Lablup will showcase significant case studies demonstrating the transformative impact of AI infrastructure across diverse fields including higher education, national research projects, and international trade.

![SC25_LABLUP_BOOTH_3102.jpg](https://cdn.lablup.com/SC_25_LABLUP_BOOTH_3102_87eaf8008c.jpg)

### Pioneering AI Solutions Across Organizations

Lablup’s commitment to making AI accessible and manageable is evident through successful implementations in various high-profile projects:

- **Korea's National AI Initiative**: In collaboration with Upstage, Lablup contributed to the 'Sovereign AI Foundation Model Project,' enabling large-scale language model development and strengthening Korea’s AI capabilities through resilient GPU infrastructure.
- **University of Southern California (USC)**: By utilizing Backend.AI’s GPU virtualization, USC supported over 80 students simultaneously in machine learning exercises, significantly enhancing AI education without the need for extensive technical staff.
- **teamreboott**: Specializing in trade automation, teamreboott fine-tuned large language models using Backend.AI FastTrack 3, streamlining trade operations with reduced human oversight.
- **Korea Institute of Science and Technology (KIST)**: KIST accelerated innovative Cas9 research using Backend.AI Cloud for dynamic molecular simulations, demonstrating the platform’s ability to support complex scientific endeavors.
- Kookmin University (KMU): KMU efficiently scaled its AI education program to accommodate hundreds of students using Backend.AI to virtualize GPU resources, demonstrating significant flexibility and broad access to technology.
- Sungkyunkwan University (SKKU): As Korea's first university to manage an independent supercomputing center, SKKU uses Backend.AI to centralize its HPC resources, optimizing research productivity and operational efficiency.

### Enhancements for Scalability and Accessibility
"Backend.AI’s self-repairing infrastructure ensures our large-scale projects remain resilient and uninterrupted,” said Jeongkyu Shin, CEO of Lablup. “Our partnerships across diverse sectors underscore our mission to eliminate AI management complexities, making advanced computing accessible to all."

### Interactive Experience at SC25
At booth #3102, Lablup will offer live demonstrations of Backend.AI running on Intel Gaudi 3 AI accelerators. Attendees can participate in interactive sessions, gain hands-on experience, and earn incentives through involvement in the Innovation Pass Program with Intel.

### About Lablup
Lablup Inc. is redefining how modern AI systems are powered by delivering the operating model for scalable, high-performance GPU computing. Its flagship platform, Backend.AI, combines software-defined GPU virtualization, ultra-fast containerized environments, and intelligent workload orchestration to maximize performance, utilization, and efficiency across heterogeneous clusters.
Founded in 2015, Lablup works with global technology leaders, research institutions, and top universities to advance AI infrastructure and accelerate data-driven innovation worldwide. Learn more at [lablup.com](https://www.lablup.com) and follow Lablup on [LinkedIn](https://www.linkedin.com/company/lablup).









Lablup Inc. to Participate in the World’s Largest HPC Event, SC25

- Lablup and VAST Data Power Korea’s Sovereign AI Initiative
- Collaboration resolves performance bottlenecks and enables large-scale model training across 500+ GPUs, advancing Korea’s national AI infrastructure.

Lablup Inc., the developer of the open, scalable AI computing platform Backend.AI and a key participant in South Korea’s government-backed Sovereign AI Project today announced a strategic collaboration with VAST Data, the AI Operating System company. Together, the companies have built a high-performance, scalable data foundation that enables Korea’s national AI consortium, which aims for technological self-reliance, to train and deploy large-language models entirely on domestic infrastructure.

Backend.AI provides a unified environment for AI model training, deployment, and inference across GPU and CPU resources. By integrating the VAST AI Operating System, Lablup and its customers can manage training data, model checkpoints, and artifacts with unmatched throughput and resilience, ensuring smooth data flows from model development to real-time serving.

As Lablup’s AI workloads grew, traditional infrastructure storage systems struggled to keep pace with the throughput and concurrency required for foundation-model training. The team turned to VAST Data to modernize its data layer, eliminating performance bottlenecks while maintaining reliability and control for multi-tenant research environments.

Running on SK Telecom’s Sovereign AI cluster, the integrated Backend.AI and VAST AI Operating System now orchestrates and accelerates model training across more than 500 NVIDIA GPUs, giving Korean researchers a unified environment for both training and, in the future, inference.

“Our mission has always been to democratize AI infrastructure through open, efficient, and scalable technology,” said Jeongkyu Shin, Founder and CEO of Lablup Inc. “By integrating VAST’s AI Operating System into Backend.AI environments, we’re enabling developers and enterprises to handle massive datasets and complex workloads with the reliability and performance that modern AI demands. It’s a technological synergy that really unlocks the full potential of our AI compute while ensuring sovereignty and data security.”

Through direct integration with the VAST AI Operating System, Backend.AI achieves seamless, high-throughput access to data for massive training workloads.Using VAST’s Disaggregated, Shared-Everything (DASE) architecture, Lablup can scale compute and storage independently while maintaining a single global namespace, VAST DataSpace, enabling researchers to access checkpoints, datasets, and model outputs instantly across projects.

Powering Lablup’s role in Korea’s Sovereign AI initiative, the Backend.AI and VAST AI Operating System deliver everything from large-scale model training to real-time orchestration, including:
- Sovereign Data Control: AI sovereignty begins and ends with data infrastructure. Built on VAST DataSpace and the AI Operating System’s Zero Trust architecture, Lablup enforces data residency, access control, and compliance across Korea’s Sovereign AI environments. For government, research, and regulated enterprise participants in the national consortium, this ensures not just where data lives, but how it’s protected, governed, and audited throughout the AI lifecycle.
- High-Performance Data Infrastructure: Backend.AI integrates with the VAST DataStore, delivering high-throughput access to training datasets and model artifacts via NVIDIA GPUDirect Storage (GDS) and RDMA. This direct GPU-to-data pipeline removes I/O bottlenecks and ensures consistent performance across more than 500 NVIDIA GPUs on SK Telecom’s Sovereign AI cluster, accelerating model pre-training and fine-tuning at national scale.
- Unified Global Namespace for Seamless Collaboration: Backend.AI, built on VAST’s DASE architecture, provides researchers and developers with a single, unified namespace that spans every workload; from model training and validation to deployment. With VAST DataSpace, teams can instantly access checkpoints, datasets, and outputs across distributed environments without duplication or downtime, simplifying collaboration across the consortium.
- Secure, Multi-Tenant Scalability: Lablup leverages VAST DataEngine and DataSpace to unify data and compute management across shared GPU clusters. Tight integration between VAST’s quota model and Backend.AI’s user-based access controls ensures secure isolation between tenants while maintaining predictable performance for every user. This architecture allows Lablup to scale compute and storage independently while providing consistent service to government, academic, and enterprise researchers alike.
- Future-Ready for Inference and Agentic AI: As the Sovereign AI initiative transitions from development to deployment, Backend.AI’s integration with the VAST AI Operating System provides a foundation for inference and future agentic AI workloads. With VAST’s DataEngine enabling real-time orchestration and reasoning across massive datasets, Lablup’s customers can move beyond static model training toward dynamic, intelligent systems that continuously learn, adapt, and serve.

Within Lablup’s own infrastructure, the VAST AI Operating System supports AI infrastructure development, optimization testing, and model service orchestration. This integration allows engineers to test and deploy new AI models more efficiently, switching model weights in real time without downtime. Through joint engineering efforts, VAST and Lablup resolved early latency challenges, resulting in a stable, high-throughput system that scales seamlessly for both training and research operations.

Within the Korean Foundation Model Consortium, led by Upstage and supported by other national AI research partners, Backend.AI powered by the VAST AI Operating System processes massive-scale corpora for large-model pre-training and fine-tuning, advancing the development of frontier-grade, sovereign AI capabilities for Korea.

“Our work with Lablup shows what modern AI infrastructure actually looks like when it’s built for scale,” said Alon Horev, Co-Founder and CTO, VAST Data. “Legacy stacks that were never designed for GPU-dense or sovereign environments simply can’t keep up. By bringing compute orchestration and data intelligence together inside the VAST AI Operating System, we’ve created a single fabric that removes those bottlenecks and gives nations like Korea the ability to own, control, and accelerate their own AI innovation. It lets teams like Lablup train bigger, move faster, and stay in control of their data. With Backend.AI and VAST Data together, enterprises and researchers are establishing a new blueprint for high-performance AI infrastructure; one that fuses flexibility, efficiency, and trust for the agentic computing era.”

Led by the Ministry of Science and ICT, Korea’s Sovereign AI initiative aims to develop globally competitive foundation models built entirely with domestic technology, reducing dependence on foreign cloud and software providers. Together, VAST and Lablup provide the data and compute fabric that makes this possible, empowering start-ups, enterprises, and research institutions to train and scale AI securely within national borders. This collaboration reflects the growing technical synergy between Backend.AI’s distributed container orchestration and the VAST AI Operating System, a unified foundation for the agentic AI era.

**About Lablup**
Lablup Inc. is redefining how modern AI systems are powered by delivering the operating model for scalable, high-performance GPU computing. Its flagship platform, Backend.AI, combines software-defined GPU virtualization, ultra-fast containerized environments, and intelligent workload orchestration to maximize performance, utilization, and efficiency across heterogeneous clusters.

Founded in 2015, Lablup works with global technology leaders, research institutions, and top universities to advance AI infrastructure and accelerate data-driven innovation worldwide. Learn more at lablup.com and follow Lablup on [LinkedIn](www.linkedin.com/company/lablup).

**About VAST Data **
VAST Data is the AI Operating System company – powering the next generation of intelligent systems with a unified software infrastructure stack that was purpose-built to unlock the full potential of AI. The VAST AI OS consolidates foundational data and compute services and agentic execution into one scalable platform, enabling organizations to deploy and facilitate communication between AI agents, reason over real-time data, and automate complex workflows at global scale. Built on VAST’s breakthrough DASE architecture – the world’s first true parallel distributed system architecture that eliminates tradeoffs between performance, scale, simplicity, and resilience – VAST has transformed its modern infrastructure into a global fabric for reasoning AI. Learn more at vastdata.com and follow VAST Data on [LinkedIn](https://www.linkedin.com/company/vast-data/), [YouTube](https://www.youtube.com/vastdata) and [X](https://x.com/VAST_Data).

Lablup and VAST Data Power Korea’s Sovereign AI Initiative with Integrated Backend.AI and the VAST AI Operating System

**SEOUL, South Korea – October 16, 2025** – Lablup Inc., a leading AI infrastructure company, today announced its participation as a silver sponsor at KubeCon + CloudNativeCon North America 2025, taking place November 10-13 at the Georgia World Congress Center in Atlanta.

The conference hosted by the Cloud Native Computing Foundation (CNCF) celebrates Kubernetes' 10th anniversary and brings together the world's largest gathering of cloud-native technology experts. Lablup demonstrates how Backend.AI, its enterprise AI infrastructure operating platform, seamlessly integrates with the Kubernetes ecosystem while addressing real-world challenges in AI workload management.

## Bridging the gap between AI innovation and infrastructure reality

"There's a harsh gap between the ideal of AI factories and reality," said Joongi Kim, CTO of Lablup. "Software is the key to bridging this gap."

At the recent 'lab | up > /conf/5' technical conference, Kim shared insights from operating large-scale GPU clusters: "When running hundreds or thousands of GPUs, something is always bound to fail somewhere." This reality drives Lablup's approach of combining horizontal scaling with vertical integration, spanning from AI models down to hardware drivers.

## Enterprise AI infrastructure operating platform 'Backend.AI' to showcase cloud-native integration

At KubeCon, Lablup will share its vision for Backend.AI as a "Container-based, Acceleration-first" platform that brings precise Kubernetes control to the cloud-native ecosystem. The platform manages heterogeneous accelerator environments through a single interface while ensuring automatic failover and recovery when individual GPUs fail, maintaining overall system stability.

Backend.AI abstracts infrastructure complexity, allowing data scientists and ML engineers to focus on models and algorithms while providing system administrators with the control and visibility needed to manage large-scale deployments. This approach reduces operational overhead and enables AI developers to concentrate fully on their core work.

## Meeting the challenges of AI as critical infrastructure

As AI infrastructure evolves into a critical industry requiring consideration of power grids and water resources, Lablup is committed to democratizing AI infrastructure through software innovation. "Backend.AI empowers the global AI community to accelerate discovery and innovate with confidence," Kim added. "We're excited to engage with the cloud-native community at KubeCon and explore collaboration opportunities."

---

## About Backend.AI

Backend.AI is an AI infrastructure operating platform that transforms GPU complexity into operational simplicity. The platform provides unified management of AI/ML workloads and HPC environments, supporting diverse languages and frameworks. The platform enables efficient resource allocation, allowing developers to easily experiment, deploy, and scale models while organizations achieve operational efficiency and cost savings.

Lablup Joins KubeCon + CloudNativeCon North America 2025 as Silver Sponsor

![DSC03604.jpg](https://cdn.lablup.com/DSC_03604_24ec8bde2f.jpg)

*   **Lablup hosts 5th annual technology conference on September 24 at the aT Center**
*   **Releases Backend.AI 25.14 with enhanced stability and greater scalability**
*   **Unveils next-generation products including AI:DOL alongside the third update of the FastTrack MLOps platform**

Lablup, a leading AI infrastructure platform company, hosted its fifth annual technology conference ‘lab | up > /conf/5’ on September 24 at the aT Center in Yangjae, Seoul. Now in its fifth year, the event served as a forum to share Lablup’s latest technological achievements and product innovations, while bringing together AI industry leaders and practitioners to discuss challenges and solutions.

Held under the theme “Make AI Composable with Lablup,” the conference featured speakers from Rebellions, Moreh, kt cloud, HyperAccel, Lotte Innovate, Miracom Inc., Microsoft, Teddynote, AIM Intelligence, SKT, Nota AI, NHN Cloud, NAVER Cloud, ETRI, and Modulab, sharing insights on advances across AI hardware, infrastructure, and software. In addition, the conference offered attendees to have demonstrations of Lablup’s latest product updates, including Backend.AI 25.14.

### Charting the path toward AGI-ready infrastructure

In the opening keynote, Jeongkyu Shin, CEO of Lablup, delivered a talk _‘_AAA: Agentic, Autonomous, Adaptive Intelligence’_._ He explained that as the AI field races toward the threshold of AGI, innovations across all layers are required to underpin this progress. Introducing the concept of “quantifying intelligence,” Shin emphasized Lablup’s new vision of transforming into a company that quantitatively supplies intelligence on demand. He also announced technological shifts spanning the Backend.AI Continuum project for fault-tolerant LLM and multimodal operations, a new App Proxy system, Backend.AI Artifact, and the AI-native platform AI:DOL.

In a parallel keynote, CTO Joongi Kim delivered a talk _‘_Composable AI, Composable Software’_,_ emphasizing that productivity gains hinge on leveraging AI models optimized for diverse purposes and applying them to the right tasks. He underscored the need to marshal the full capabilities of the computing industry from physical hardware to abstraction and management software to AI models into a cohesive, composable ecosystem.

### AI:DOL – A next-generation platform for natural language AI design and prototyping

AI:DOL is Lablup’s new AI-native platform for enterprises and individual developers aiming to provide AI services to both internal and external clients using models deployed on Backend.AI. Built on Backend.AI Continuum, AI:DOL integrates on-premises and cloud resources with seamless failover, enabling users from beginners to experts to easily prototype and create.

Hyounkyoung Moon, who led the development, showcased AI:DOL in the session ‘Building Conversational AI Products with Backend.AI Continuum’, demonstrating its chat-based interface for designing and deploying AI with natural language. Through conversational interactions, users can generate documents, code, and other artifacts, which is also planned to have expansion into image, video, and AI coding capabilities.

### Backend.AI FastTrack 3 – Workflow automation on hyperscale GPU clusters

The updated Backend.AI FastTrack 3 platform was introduced as a workflow engine running atop Backend.AI, which unifies massive GPU resources into a single pool to power the training and serving of frontier-scale models.

Eunjin Hwang and Jeongseok Kang presented ‘Take the FastTrack 3: A Backend.AI approach to LLMOps’, introducing an independent AI foundation model project aimed at developing a Korean-language model reflecting the nation’s cultural identity. They demonstrated how Backend.AI and FastTrack 3 contribute to efficient model development through workflow orchestration and large-scale GPU utilization.

### Backend.AI – The center of cloud-native infrastructure

At the core of Lablup’s portfolio, Backend.AI is an AI infrastructure operating platform built for every scale. Backend.AI version 25.14 has been engineered through Lablup’s ongoing ‘hardening’ process to deliver stronger stability and support AI workloads at unprecedented scale across distributed environments.

### Composable AI – unlocking new possibilities

_“Lablup has been driving innovation in distributed AI systems and pursuing transformative changes since 2024,” said Jeongkyu Shin, CEO of Lablup. “With the unveiling of AI:DOL, Backend.AI FastTrack 3, and Backend.AI 25.14, we reaffirm our vision as a platform technology company driving a better AI-powered future.”_

lab | up > /conf/5 highlighted the importance of composability in AI. By emphasizing modular AI architectures that can be selected and combined as needed, the conference showcased how independently evolving AI components can operate as a single ecosystem, unlocking new possibilities for innovation.

Lablup Hosts lab | up > /conf/5, Presenting the “Make AI Composable” Vision

![[래블업] AI Infra Summit 2025_AIDOL-F.png](https://cdn.lablup.com/AI_Infra_Summit_2025_AIDOL_F_6457ca839a.png)

Seoul, South Korea – August 29, 2025 – Lablup, Inc. announced today that it will participate in AI Infra Summit 2025, showcasing its comprehensive AI infrastructure management capabilities. 

AI Infra Summit 2025 brings together over 3,500 global AI professionals and 100+ partner companies to address the full spectrum of AI infrastructure from hardware to data centers and edge environments. Lablup will demonstrate at booth #620 the beta version of its AI-native unified platform AI:DOL alongside Backend.AI running on a miniature AI data center built with palm-sized NVIDIA Jetson Orin Nano devices. 

Lablup unveils the beta version of its AI-native platform 'AI:DOL (AI Deployable Omnimedia Lab)'. AI:DOL combines Backend.AI Core with PALI-based services to deliver a generative AI development platform accessible via web browser without installation. It supports developers from beginners to experts, enabling AI model deployment in a low-code environment. Through Backend.AI Continuum, AI:DOL integrates on-premises and cloud resources, enabling streamlined development and deployment of AI applications for internal and external clients. Currently focused on chat functionality, AI:DOL plans to expand into image, video, and AI coding capabilities to become a comprehensive AI development ecosystem. 

Lablup also showcases the 'Mini AI Factory' and demonstrates how the same Backend.AI platform scales from edge devices to hyperscale data centers with over 1,500 GPUs. Visitors can witness modular scalable architecture, heterogeneous AI accelerator management, and patented container-level GPU virtualization technology. Showcase highlights the infrastructure requirements for Sovereign AI and the importance of compliant, efficient infrastructure management solutions. 

Lablup participates in a South Korean government-led AI foundation model project managed by the Ministry of Science and ICT as part of the Upstage consortium. The company operates NVIDIA B200 GPU infrastructure in SK Telecom's Haein cluster and collaborates with Upstage, NVIDIA, and SK Telecom to research stable and accelerated large-scale foundation model training technologies. This involvement supports Korea's national Sovereign AI initiative while strengthening Lablup's global competitiveness through large-scale AI training technology advancement. 

_"The U.S. market is experiencing unprecedented demand for AI infrastructure and managed backend platforms that align with rapidly growing AI workloads," said Jeongkyu Shin, CEO of Lablup, Inc. "Growing focus on Sovereign AI further accelerates AI infrastructure demand. Yet, scaling AI reliably across diverse and distributed environments remains a major challenge. Backend.AI addresses this need with a globally deployable, scalable solution that enables confident AI expansion at every stage."_ 

Lablup aims to strengthen its global presence while highlighting its role in shaping the next generation of enterprise AI infrastructure.  

**About Lablup** 

Lablup, Inc. develops Backend.AI, an AI infrastructure operating platform built for every scale. Continuously building a product that makes it easy to operate and scale AI. We strive to address the challenges of sharing scientific and engineering research through a standardized platform, while also automating the entire process—from preparing research to delivering final code. With its innovative technology, Lablup enables the global AI community to accelerate discovery and innovate with confidence. 

**Media Contact** 

DevRel Team | Lablup, Inc.   
Representative Email: [wyyoo@lablup.com](mailto:wyyoo@lablup.com)   
Website: [www.lablup.com](http://www.lablup.com/) | [www.backend.ai](http://www.backend.ai/)

Lablup Unveils AI-Native Platform 'AI:DOL' at AI Infra Summit 2025, Scaling from Edge Devices to Hyperscale Data Centers & National AI Projects 

![ISC2025.jpg](https://cdn.lablup.com/ISC_2025_9a09a64c5b.jpg)

Seoul, South Korea – Lablup Inc. will participate in ISC High Performance 2025, the world’s largest high-performance computing (HPC) conference, held from June 9 to 11 in Hamburg, Germany. At Booth C52, Lablup will present its flagship product, Backend.AI Enterprise, as a key driver for its expansion into the European AI infrastructure software market. 

During the event, Lablup will demonstrate Backend.AI running on NVIDIA Jetson Nano micro-clusters, highlighting the platform’s flexibility and scalability in efficiently managing infrastructure from edge computing environments to large-scale data centers. The company will showcase live AI model training and inference workflows, emphasizing how Backend.AI leverages container-level GPU virtualization and other features to maximize the use of limited hardware resources while ensuring stable and reliable AI development cycles. 

On June 10, Lablup’s Director of Research, Jonghyun Park, will deliver a keynote introducing the company’s products and vision. The presentation will emphasize Lablup’s technical expertise and its ambitions within the global AI market. 

Throughout ISC 2025, Lablup plans to engage in meetings with numerous European technology and sales partners, as well as research institutions, with the aim of establishing new partnerships and further solidifying its presence in the European AI infrastructure software market. 

Backend.AI is designed to automatically distribute and manage GPU resources across tens to thousands of devices, enabling developers to rapidly conduct AI and HPC research and development without the need for complex environment setup or resource optimization. The platform is compatible with cloud, on-premises, hybrid, and air-gapped environments, making it adaptable to a wide range of infrastructure needs. Backend.AI is already deployed at scale in major enterprises and institutions worldwide, including Samsung Electronics, KT, and Hyundai Mobis, and is recognized as an innovative solution in the rapidly evolving global AI and HPC landscape. 

“Backend.AI Enterprise, already proven by diverse customers both domestically and internationally, is set to enter the European market starting with ISC 2025,” said Jeongkyu Shin, CEO of Lablup. “We are committed to spreading our philosophy of AI democratization worldwide. With strong demand for HPC in manufacturing and research sectors, we expect significant synergies, especially in Germany and across Europe.”

Lablup Showcases Backend.AI Enterprise at ISC 2025 in Germany, Expanding into the Global HPC Software Market

**\- The company's first North American exhibition since incorporating in the U.S., aims to take on the global AI infrastructure market**

![gtc25-social-drmk-orangic-exhibitor-reg-now-li-1200x628.jpg](https://cdn.lablup.com/gtc25_social_drmk_orangic_exhibitor_reg_now_li_1200x628_7dbc9fe7d1.jpg)

\[Mar 18, 2024\] Lablup, the AI infrastructure operating platform company, announced it will showcase new AI projects at NVIDIA GTC 2025, which runs March. 17-21 in, San Jose, California. The company will showcase its cloud fault tolerance solution, 'Backend.AI Continuum', along with the 'Backend.AI v25.3' update, and 'Backend.AI for Personal Supercomputer'.

Lablup has been preparing to go global since 2023. Lablup already established a U.S. subsidiary, 'Lablup USA Inc.' earlier this year, and marks its entry into the North American market. GTC 2025 will be Lablup's first North American exhibition since its establishment, and the company will showcase its core products to target the global AI infrastructure market. 

“We are excited to introduce Backend.AI Continuum and Backend.AI for Personal Supercomputers at GTC 2025,” said Jeongkyu Shin, CEO of Lablup. “Backend.AI Continuum acts as a UPS (Uninterruptible Power Supply) to ensure continuity of AI services in the event of cloud API outages, and Backend.AI for Personal Supercomputers will help democratize AI by leveraging NVIDIA's latest edge computing platforms. These two innovations will position Lablup as the global leader in providing a seamless AI infrastructure experience from cloud to edge.”

“Establishing a U.S. subsidiary is the first step in our global expansion strategy, and we plan to expand our business from North America to Europe and Asia Pacific,” said Shin. “We are confident that our technology and innovation will be recognized in the global market with our proven AI infrastructure.”

'Backend.AI Continuum' is an innovative solution that enables organizations using cloud API-based services to continue their business even in the event of network failures or service outages. The technology intelligently bridges the gap between the cloud and on-premises environments during normal operation and automatically switches to local resources in the event of a cloud access failure, keeping API calls and processing uninterrupted. This gives customers the flexibility of the cloud and the reliability of on-premises at the same time.

As global organizations increasingly rely on cloud services, Backend.AI Continuum is expected to be a key product in the North American market. Lablup is currently working on pilot projects with several potential customers in the U.S. and is looking forward to rapid market expansion, particularly in the financial and healthcare industries.

At GTC 2025, Lablup is also showcasing Backend.AI for personal supercomputers powered by NVIDIA accelerated computing. Running on personal supercomputer products such as NVIDIA DGX Spark or edge computing devices based on Jetson Orin™ Nano series™ modules, the lightweight version of Backend.AI optimizes the core functionality of the original Backend.AI for use on consumer desktops, delivering high-density generative AI performance on edge AI computing environments.

'Backend.AI for Personal Supercomputer' is specialized in scalability and performance, with a user-friendly installer and WebUI Neo interface. This makes it easy to install and use for both beginners and experts. Lablup plans to proactively target the individual developer and SMB market in North America with Backend.AI for Personal Supercomputer.

During GTC 2025, Lablup will be hosting live demos of Backend.AI Continuum and Backend.AI for Personal Supercomputer at its booth (#547), as well as a special session to present examples of integration within the NVIDIA ecosystem. Lablup will also be hosting a number of private meetings with potential customers and partners in the US.

Lablup Unveils 'Backend.AI Continuum' and 'Backend.AI for Personal Supercomputer' at NVIDIA GTC 2025

![Lablup_KT MOU-01.jpg](https://cdn.lablup.com/Lablup_KT_MOU_01_9b3a9ab2ac.jpg)

Original: [KT Partners with Lablup for GPU Subscription Service... "Expanding AI Ecosystem"](https://www.yna.co.kr/view/AKR20250227043700017)

KT \[030200\] announced on the 27th that it will pursue Graphics Processing Unit (GPU) cloud service (GPUaaS) business with AI startup Lablup.

The two companies agreed to jointly develop GPUaaS services, which have seen increased demand in the recent global AI transformation, and launch the service as soon as possible.

Lablup is a company that develops AI infrastructure operation management platforms. Its flagship product, Backend.AI solution, is the only software in the Asia-Pacific region to receive verification for use with NVIDIA DGX systems.

KT explained that Lablup is the optimal partner to jointly pursue the GPUaaS service business, which lends GPU resources to companies requiring high-performance computing.

Jeongkyu Shin, CEO of Lablup, said, "We will support corporate customers' AI transformation by combining our technology, which provides AI solutions from the operating system level to services, with KT's expertise."

Seung-ho Song, head of KT's Strategy and Business Consulting Division's SPA Headquarters, said, "We aim to collaborate with multiple partners for solutions that help corporate customers' AI transformation," adding, "We will work with Lablup to support GPUaaS services to many companies that need GPUs."

  
csm@yna.co.kr

KT Partners with Lablup for GPU Subscription Service... "Expanding AI Ecosystem"

> \- Demonstrates technology securing AI service continuity through automatic local switching during cloud connection failures  
> \- Significantly enhances service stability in mission-critical areas including finance, healthcare, and public safety

AI infrastructure platform company Lablup announced its participation in the world's largest mobile communications exhibition, Mobile World Congress (MWC) 25, taking place in Barcelona, Spain from March 3 to 6, marking its third consecutive year. At this exhibition, Lablup will unveil the next-generation version (v25.3) of its flagship product Backend.AI, highlighting 'distributed fault tolerance' technology that ensures cloud API service continuity as a core innovation.

The new distributed fault tolerance technology incorporated in Backend.AI v25.3 enables enterprises using cloud API-based services to continue critical operations despite network failures or service interruptions. This technology normally serves as an intelligent intermediary between cloud and on-premises environments, but automatically switches to local resources during cloud connection failures to maintain API calls and processing without interruption. This allows organizations to secure both the flexibility of cloud and the stability of on-premises simultaneously.

The value of this technology shines even more in environments like smart cities. Even when cloud API access becomes impossible due to internet connection failures, traffic control, security monitoring, and emergency response systems continue to operate uninterrupted in the local environment. This represents an innovation that dramatically reduces recovery time and greatly increases service stability compared to existing cloud-only solutions.

The distributed fault tolerance technology of Backend.AI is expected to deliver significant value in areas where service interruptions are critical, including ▲financial transaction systems ▲medical monitoring ▲public safety sectors. Lablup has confirmed through internal testing that this technology substantially improves system stability and is currently conducting validations with several companies.

Lablup will conduct demonstrations simulating real-time cloud connection failures at **Hall 3 (K-AI Alliance Exhibition Hall)** and **Hall 7 (#7A62)**. Visitors can directly experience how Backend.AI immediately switches to a local environment to maintain service operations when cloud service connections are lost.

Additionally, the deep learning model deployment solution 'PALI (Performant AI Launcher for Inference)' will be showcased. Released in September last year, PALI has been combined with Backend.AI's distributed fault tolerance technology to further enhance the stability of inference services.

"The new distributed fault tolerance technology in Backend.AI v25.3 is an innovation that addresses the fundamental issue of instability due to cloud service dependency," said Jeongkyu Shin, CEO of Lablup. "This technology, which intelligently links cloud and local environments, has opened the way for confidently utilizing cloud AI technology even in mission-critical industries." He added, "Through this MWC, we plan to introduce the new value of Backend.AI v25.3 to the global market and expand domestic and international partnerships in the future."

Lablup Unveils New Backend.AI Technology Ensuring Cloud API Service Continuity at MWC 2025

![PALI-demo-screenshot-1.png](https://cdn.lablup.com/PALI_demo_screenshot_1_2ceed51650.png)

Lablup has announced participation in CES 2025 scheduled for January 7-10 in Las Vegas USA. CES is the world's largest IT and consumer electronics trade show, featuring more than 4,300 companies from 158 countries around the world. Lablup, which has participated in CES for three consecutive years since 2023, will have a booth at LVCC, North Hall, #9528, and will showcase PALI (Performant AI Launcher for Inference), a total inference solution that makes it easy and fast to serve deep learning models.

PALI's key advantages include the following: Support for deep learning models from various sources such as NVIDIA NIM, Hugging Face, Kaggle, etc. Run the latest AI models such as language models and image generation models easily like apps. Easily deploy or operate the system even in on-premises and air-gapped environments. Powerful Backend.AI technology which maximizes NVIDIA GPU hardware performance. Intuitive, convenient user and administrator interface. Flexible scalability based on compliance hardware so that customers can quickly apply the latest AI models to their business.

Already available in the Japanese market, the PALI suite is ideal for organizations looking to quickly adopt the latest AI models into their business. With PALI, developers can focus on developing AI services without the complex infrastructure setups required to serve models and can easily implement high-performance AI services for large-scale users on top of the Backend.AI platform with outstanding GPU acceleration performance and flexible scalability.

“With the advancement of AI technology, the demand for AI services and applications in various scales and fields is rapidly increasing,” said Jeongkyu Shin, CEO of Lablup. ”Based on Lablup's proven Backend.AI technology, we will actively support the success of AI projects by providing the best platform solution in the field of high-performance and high-efficiency inference. Lablup will continue to grow as an industry-leading AI specialist by developing cutting-edge AI technologies and advancing its products.”

In addition to PALI, Lablup will also present its Hyperscale AI Software Platform, Backend.AI. Backend.AI is a vendor-agnostic accelerated workload hosting platform based on our own home-grown orchestration and job scheduler, running on top of either cloud or on-premises (air-gapped) clusters. Backend.AI currently has more than 70 enterprise customers both domestically and internationally, including Brazil and the United Kingdom, with customers across a range of industries including manufacturing, finance, defense, IT, healthcare, and universities.

![PALI-demo-screenshot-2.png](https://cdn.lablup.com/PALI_demo_screenshot_2_766d7ef839.png)

![PALI-demo-screenshot-3.png](https://cdn.lablup.com/PALI_demo_screenshot_3_1cf6194685.png)

![PALI-demo-screenshot-4.png](https://cdn.lablup.com/PALI_demo_screenshot_4_af331e36de.png)

Lablup to unveil PALI, a new product that highlights AI scalability at CES 2025

![lablup-sc24](https://cdn.lablup.com/lablup_sc24_719c153b8c.jpg)

Lablup, an all-in-one enterprise development and service platform company for artificial intelligence development, announced its participation in Supercomputing 2024 (SC24), held in Atlanta, USA, from November 17 to 23.

SC24 is the largest supercomputing conference, hosted annually since 1988 by the Association for Computing Machinery (ACM) and IEEE Computer Society, where the latest research and technology in high-performance computing (HPC), networking, storage, and analytics are shared.

Participating in the exhibition for the second consecutive year, Lablup will unveil version 24.12 of its AI infrastructure platform Backend.AI at this event. Additionally, they will showcase demos of PALI, an inference solution based on Backend.AI 24.12, optimized for various AI semiconductors. Since NVIDIA's Grace CPU was unveiled in 2021, Lablup promptly released a new Backend.AI version compatible with the Grace architecture, completed verification for the GH200 Grace Hopper Superchip, and prepared for NVIDIA's Blackwell platform to be released in 2024. Furthermore, through a six-month collaboration partnership with Intel, they optimized Backend.AI to maximize the performance of Intel Gaudi 2 and Intel Gaudi 3 AI accelerator platforms.

At SC24, Lablup will demonstrate AI model operations and development on Intel Gaudi 2 and Intel Gaudi 3 AI accelerator platforms and NVIDIA Grace Hopper Superchips using PALI. Simultaneously, through various booth sessions with partners, they will introduce the foundational technologies for operating, optimizing, and managing large-scale inference services with PALI.

CEO Jeongkyu Shin stated, "Lablup is a company that accelerates computing. The AI field, which demands massive computing power, is becoming closely integrated with supercomputing," adding, "At SC24, where the frontiers of computing including AI, accelerated computing, storage, networks, and quantum computers unfold, we will share Lablup's advanced technology with the world."

Lablup Announces Enhanced Large-scale GPU and AI Accelerator Cluster Management Capabilities at SC24

![webui-neo](https://cdn.lablup.com/webui_neo_a6025850d5.jpg)  
Lablup announced to participate in KubeCon + CloudNativeCon North America 2024, which will be held in Salt Lake City from November 12 to 15. At this event, Lablup will showcase the latest version (24.09) of Backend.AI, along with its new journey toward a user-friendly frontend, WebUI Neo.

Backend.AI 24.09 was developed with the aim of managing ultra-large-scale GPU and AI accelerator clusters and enhancing system visibility through integration with Cloud Native. Starting with this version, Backend.AI supports Prometheus-format metric outputs, enabling users to easily connect various metrics managed by Backend.AI to dashboard solutions such as Grafana.

With 24.09, Backend.AI addresses limitations in the existing Docker network stack by adopting Cilium. This integration reduces dependency on Docker, improves the stability and throughput of container-to-container networks, and meets the demands of various container runtimes in the market.

Backend.AI is set to integrate fine-grained Kubernetes control by 2025, evolving into an acceleration-first, container-based AI infrastructure operating system that fully embraces the Cloud Native ecosystem.

WebUI Neo reimagines Backend.AI, traditionally function-focused, with a user-centered approach. It features an interface that enhances user feedback, clearly representing statuses through color, making Backend.AI’s wide array of functionalities more accessible in a streamlined, logical interface. WebUI Neo is expected to be released within 2024.

At Lablup's booth R27, visitors can explore Backend.AI, the acceleration-first, container-based AI infrastructure operating system that fully integrates with the Cloud Native ecosystem.

"At KubeCon, Lablup will share the present and future of Backend.AI, the potential of its Kubernetes compatibility, and the enhanced user experience offered through WebUI Neo," said Jeongkyu Shin, CEO of Lablup. "We invite you to see firsthand how Backend.AI can efficiently manage your AI clusters."

![Backend.AI package(Kubecon 2024)](https://cdn.lablup.com/Backend_AI_package_Kubecon_2024_4f643a6656.jpg)

Lablup Unveils Backend.AI 24.09 and WebUI Neo at KubeCon

![nvidia-ai-summit-jp-speakers](https://cdn.lablup.com/nvidia_ai_summit_jp_speakers_bb41639d48.png)

AI platform company Lablup has announced its participation in the '[NVIDIA AI Summit](https://www.nvidia.com/ja-jp/events/ai-summit/fireside-chat-en/)' held in Tokyo, Japan. The NVIDIA AI Summit introduces the latest trends in AI and accelerated computing technologies along with real-world use cases, bringing together experts and industry leaders from various fields including science, healthcare, and sustainability to discuss AI's future and impact. This year, the summit is being held in Washington D.C., India, and finally in Japan.

At this NVIDIA AI Summit Japan, NVIDIA founder and CEO Jensen Huang and SoftBank Chairman Masayoshi Son will engage in a dialogue about how AI and accelerated computing will transform industries. The event, which has received over 3,000 registrations, begins with workshops on the 12th and features more than 50 diverse sessions and live demos covering topics such as generative AI, industrial digitalization, robotics, and Large Language Models (LLM).

Among 62 companies, Lablup will demonstrate how to effectively utilize AI technology through NIM (NVIDIA Inference Microservices) by running its hyperscale AI platform, Backend.AI, and inference solution PALI on NVIDIA GH200 Grace Hopper Superchips.

Backend.AI, which has been certified as NVIDIA DGX-Ready Software since 2021 - the first and only one in the Asia-Pacific region - is an AI platform that optimizes and manages computing resources for large-scale AI and accelerated computing applications, from DGX system clusters to the latest NVIDIA Grace CPU Superchips, while enhancing GPU usability.

"Through this NVIDIA AI Summit, Lablup will showcase how we work with NVIDIA to maximize the efficiency of generative AI service operations with PALI and PALANG, as well as developing large-scale AI models based on Backend.AI," said Jeongkyu Shin, CEO of Lablup. "We aim to widely promote Backend.AI's strengths as an AI infrastructure operating system that addresses various scales, from PC and small-scale organizations' AI utilization to GPU hyperscaler AI service operations."

![nvidia-ai-summit-jp](https://cdn.lablup.com/2024_11_10_8_54_18_93c5d55e2e.png)

Lablup Participates in NVIDIA AI Summit Japan

 ![](https://cdn.lablup.com/lablup_intel_6f980e65b3.jpg) 

**Seoul —** Lablup announces support for Intel® Gaudi® 2 & Intel® Gaudi® 3 AI Accelerators* in Backend.AI at Supercomputing 2024. Adding support for Intel's AI accelerators to the list of AI accelerator vendors already supported by Backend.AI, including NVIDIA, Rebellions, FuriosaAI, AMD, and others, Lablup is able to offer its customers the most AI accelerators and GPUs on the market, making the Backend.AI platform more competitive and giving customers more choice.

*As of November 2024, Backend.AI supports the Intel® Gaudi® 2 AI accelerator. 

*Support for Intel® Gaudi® 3 AI Accelerators is planned on the first half of 2025.


Lablup and Intel have worked closely to unlock the power of Intel® Gaudi® 2 & 3 Platform and make it available in Backend.AI, and as the result of this collaboration, we are pleased to announce Backend.AI now supports the Intel® Gaudi® 2 and Intel® Gaudi® 3 AI Accelerators.

### Powerful container orchestration with Sokovan™ by Backend.AI
Sokovan™ is a standalone open-source container orchestrator highly optimized for latest hardware acceleration technologies used in multi-tenant and multi-node scaling scenarios. With fully customizable job scheduling and node allocation policies, it accommodates a hybrid mix of interactive, batch, and service workloads in a single cluster without compromising the performance of the AI.

### Get the most out of the AI Accelerators, reach your maximum possibilities.
In complex business, promising AI performance and manageability is the key to success. Intel® Gaudi® 3 AI accelerator, the latest release from Intel, offers powerful AI performance and features. Lablup Backend.AI, a Platform-as-a-Service, offers a wide range of features which are optimized for enterprise-grade AI environments.

### Innovation deeply integrated with Intel® Gaudi® 2 & 3 Platform
Customers who have already adopted Intel® Gaudi® 2 AI Accelerators or Intel® Gaudi® 3 AI Accelerators in their business environments, as well as those who will adopt Intel® Gaudi® 2 & 3 Platform in the future, will benefit from the wide range of features that Lablup Backend.AI supports for Intel® Gaudi®. Check out the following examples made possible with Backend.AI with Intel® Gaudi® 2 & 3 Platform.

#### Card-level accelerator allocation
Maximize Intel® Gaudi® 2 & 3 AI Accelerators cluster workload by lending users the actual number of accelerators intended to. For example, customers can run and train models on their existing preferred platform, then serve on Intel® Gaudi® 2 & 3 Platforms, or vice versa.
#### External storage allocation 
Get the most of out the integrated storage solutions in terms of performance. Utilize vendor-specific filesystem acceleration features without user intervention. Backend.AI supports major, widely used platforms such as Dell PowerScale, VAST Data, WEKA, NetApp, etc.,
#### Multi-scale workloads
Whatever your environment is, from single-card AI workload which can run small models, to multi-node multi-card AI workload which can run gigantic models, Backend.AI ensure its best performance. At this point as Nov.1, Backend.AI is ready to run Single-card AI workloads and Single-node, Multi-card AI workloads. Multi-node, Multi-card AI workload support will be finalized this year. 
#### Inference statistics management
Monitor up-to-date, detailed metrics about the performance provided by your AI framework. Backend.AI makes inference statistics management easy, not only showing the information from the hardware, but also on software so that administrator can deep-dive into the metrics.
#### Rule-based inference replica auto scaling 
Let the system self-optimize the resource usage.  With varied user traffic to inference workloads based on a combination of hardware and software performance metrics, administrators do not need to manually control remaining resources.

*Currently in development (Targeting Dec. 2024)

#### NUMA-aware resource allocation 
Achieve the maximum bare-metal performance by eliminating inter-CPU and PCIe bus overheads within a single node, when there are multiple CPU sockets and multiple accelerators for each socket. 

#### User-based, Project-based storage quota management 
Budget-efficient, easy data-space management by limiting data storage quota per user or single project. 

#### Hugepage memory allocation support 
Minimize the CPU overheads when using accelerators by using larger memory pages but fewer in number to reduce address translation overheads. This support will be finalized this year. 

#### ... and much more
Lablup continues to communicate with Intel, expanding the possibility of Backend.AI. Many more are still in development and will be announced soon. Elevate what your cluster can do with Backand.AI and Intel® Gaudi® 3 AI Accelerators.

### Getting most out of your Intel® Gaudi® 3 AI Accelerators.
 ![](https://cdn.lablup.com/2024_11_06_141836_9836bea9c4.png) 
Backend.AI is designed to bring out the maximum performance Intel® Gaudi® 3 AI Accelerators are capable of. Built on the high-efficiency Intel® Gaudi® platform with proven MLPerf benchmark performance, Intel® Gaudi® 3 AI accelerators are built to handle demanding training and inference. Support AI applications like Large Language Models, Multi-Modal Models and Enterprise RAG in your data center or in the cloud—from node to mega cluster, all running on the Ethernet infrastructure you likely already own. Whether you need a single accelerator or thousands, Intel® Gaudi® 3 can play a pivotal role in your AI success.


### Do these with superior user interface.
Unlike other systems, Backend.AI is designed for system administrators to control their system easily as possible. Our consumer-grade user interface makes administrators manage their system within a few clicks and types. Backend.AI WebUI is widely used by our proven customers, and they love what they can do without opening the Command Line Interface. 

### Make your Intel® Gaudi® 2 & 3 Platform 'manageable' with Backend.AI, unleash your performance.
We are making AI services efficient, scalable, and accessible to scientists, researchers, DevOps, enterprises, and AI enthusiasts. Lablup and Intel are working closely together to enable the success of Generative AI and deep learning-based services that are popular today. 
With our proven technology, Backend.AI provides hardware-level integration with Intel® Gaudi® 2& 3 Platform for the best effort.


### About Intel® Gaudi® 3 AI accelerator
Intel® Gaudi® 3 AI accelerator is driving improved deep learning price-performance and operational efficiency for training and running state-of-the-art models, from the largest language and multi-modal models to more basic computer vision and NLP models. Designed for efficient scalability—whether in the cloud or in your data center, Intel® Gaudi® 3 AI Accelerators bring the AI industry the choice it needs—now more than ever.
To learn more about Intel® Gaudi® 3, visit [intel.com/gaudi3](https://intel.com/gaudi3)

### About Lablup Backend.AI
Backend.AI supports a wide range of GPUs and AI accelerators on the market to achieve maximum efficiency of its performance and provides a user interface to make everything easy. This allows customers to efficiently build, train, and deliver AI models, from the smallest to the largest language models, significantly reducing the cost and complexity of developing and operating services. Backend.AI is the key to unlock the full potential of Generative AI and Accelerated Computing, transforming your business with cutting-edge technology.
To learn more about Backend.AI®, visit [backend.ai](https://www.backend.ai/)


Lablup - Intel Announce Support for Intel® Gaudi® 2 & 3 Platforms in Backend.AI

![lablup-conf-4](https://cdn.lablup.com/lablup_conf_4_a3b94c6250.avif)

AI platform company Lablup announced that it will hold its regular technical conference 'lab | up > conf/4' at the aT Center in Yangjae on the 24th. AI experts and corporate representatives will attend the event to provide valuable presentations to participants.

While the previous three 'lab | up > conf' events focused on 'Make AI Accessible,' discussing technological trends in improving AI accessibility, this year's theme is 'Make AI Scalable,' addressing the technologies and challenges of implementing and operating AI in large-scale industries.

At this conference, Lablup CEO Jeongkyu Shin will deliver a keynote speech on ▲'Unchartered AI: The Great AI Era.' He plans to share Lablup's experiences and future vision, along with in-depth discussions about AI foundational technologies in an era where scale bands are expanding both upward and downward. Lablup's Principal Researcher Eunjin Hwang will present ▲'I'm Getting Dizzy!: PALI, PALI PALI, and PALANG,' introducing Lablup's new platform, AI runtime cluster, and multimodal language model suite, while NVIDIA Director Chanran Kim will introduce NVIDIA NIM in ▲'NVIDIA's AI Inference Microservices for Enhancing AI Product Competitiveness.'

Additionally, Rebellion Software Engineer Jongho Choi will present ▲'RBLN SDK: A Tool for Easy Transition to Rebellion AI Accelerator,' introducing the ATOM NPU, and kt cloud AI Platform Team Leader Sungjin Hwang will discuss ▲'kt cloud's AI Business Growth Story: This is Why Partners are Important!' sharing kt cloud's growth story with Lablup. AICA Infrastructure Development Director Jaedo Kwak will also attend to share ▲'Planting AI Seeds in Korea! AI-Centered Industrial Convergence Cluster Development Project,' discussing the progress and major support cases of the Gwangju AI Industrial Convergence AI-Valley development project.

The conference will feature speakers from various IT technology companies including Rebellions, Microsoft, NVIDIA, Kakao, and kt cloud, with a total of 19 presentations including the keynote speech.

After the conference, participants will receive commemorative gifts, and various prizes including the Apple Watch Series 10 will be given away through a lucky draw.  
Lablup CEO Jeongkyu Shin stated, "We are very pleased to provide a venue where various IT technology companies can gather to share the latest technology trends and innovative ideas through 'lab | up > conf/4'," adding that "Lablup will continue to lead in the great AI era as a company that can handle scale."

'lab | up > conf/4' registration is available until today (23rd) through pre-registration on Festa ( https://festa.io/events/5923 ).

Lablup Holds Annual Technical Conference 'lab | up > /conf/4'

 ![](https://cdn.lablup.com/Lablup_GTC_24_a1eb755914.png) 

[Mar 4, 2024] Lablup is set to participate as a silver sponsor at NVIDIA's premier AI conference, GTC 2024, taking place from March 18th to 21st at the San Jose Convention Center, USA. This in-person event, the first of its kind in five years, is anticipated to be the largest tech gathering in the GPU sector, with over 300,000 participants joining both online and offline.

At the conference, Lablup will present on two cutting-edge topics within AI technology. The first presentation, titled "From Idea To Crowd: Manipulating Local LLMs At Scale," will be conducted by Lablup's CEO, Jeongkyu Shin, and CTO, Joongi Kim. This session will introduce how Backend.AI leverages resource-saving and model compression technologies such as quantization and QLoRA, along with the integration of vLLM and TensorRT-LLM, to efficiently manipulate large-scale local LLMs. This approach aims to optimize AI model performance while minimizing resource usage.

The second presentation, 'Personalized Generative AI: Operation and Fine-Tuning in Household Form Factors,' scheduled for March 20th, will be led by CTO Joongi Kim. It will demonstrate how to develop and service customized generative AI using cloud and PC GPUs, presenting new possibilities for integrating AI technology more closely into everyday life through personalized AI.

Furthermore, Lablup will showcase Backend.AI Enterprise at booth #1233, the only NVIDIA DGX-Ready software in the APAC region. Backend.AI is an AI infrastructure operation platform that maximizes the performance of NVIDIA DGX systems and other GPU-accelerated computing resources, enhancing GPU usability without altering the user environment. At the booth, Lablup will showcase FastTrack, an MLOps platform that streamlines and automates AI development, customizing foundation models for various industries and enabling real-time fine-tuning for personalized chatbots.

Following MWC, Lablup aims to strengthen cooperation with global partners such as NVIDIA, Dell, AMAX, and Weka at GTC, paving the way for a significant global expansion.

Lablup to Showcase the Future of AI Innovation at NVIDIA GTC 2024

 ![](https://cdn.lablup.com/MWC_Amax_744f5049a8.jpeg) 

[Barcelona, Feb. 26, 2024] Lablup announced on the 26th that it has held a ceremony to sign an agreement with global HPC solution company 'AMAX' at 'MWC 2024', the world's largest mobile exhibition held in Barcelona, Spain. The agreement aims to strengthen business development cooperation and expand business opportunities between the two companies.

The ceremony was attended by Lablup CEO Jeongkyu Shin and AMAX Senior Vice President Bohr Tsao. Lablup, a company specializing in AI research, development, and service platforms, and AMAX, an HPC solution company, plan to accelerate the construction of AI supercomputing infrastructure and provide GPU integrated solutions to research institutes and data centers worldwide.

The two companies have been discussing collaboration since co-hosting the "Optimising AI Conference" in London last November. At this event, technical experts and business leaders in the AI field, along with Lablup, presented cases across the AI industry, including telecommunications, healthcare, and finance, based on Lablup's UK market entry case, and had networking time.

Bohr Tsao, Senior Vice President of AMAX, said, "The current problem with IDC is not only power consumption but also heat control. We have introduced a configuration that installs a separate liquid cooling rack along with a calculation rack, which allows us to build the optimal heat control infrastructure even in the current IDC environment." He added, "We look forward to supplying a global AI infrastructure platform by combining AMAX's infrastructure design and operation experience with Lablup's Backend.AI."

Jeongkyu Shin, CEO of Lablup, said, "Lablup, which provides the world's best platform for increasingly complex AI infrastructure operation and optimization, and AMAX, which provides high-performance computing hardware platforms of various scales, are expected to create synergy." He expressed his ambition, saying, "Through this agreement with AMAX, we will provide an AI total platform targeting the high-performance computing demand in Europe and North America, allowing more people around the world to enjoy AI innovation through Backend.AI more closely."

Lablup Signs Global Agreement with AMAX for AI Infrastructure Platform Supply

![](https://cdn.lablup.com/Gen_AI_Dasktop_01_02538c826c.png)

Lablup announced its participation in the world's largest mobile exhibition, 'MWC 2024', held in Barcelona, Spain, from February 26th to 29th.

The major topic of this year's MWC 2024 is AI. Telecommunication companies are eagerly announcing their investments in the latest AI-related technologies with the aim of joining the global trend of the edge cloud market. Lablup will introduce how telecommunication companies can easily, quickly, and cost-effectively research, develop, and serve AI on the edge clouds, which are small computational resources set up at base stations. Furthermore, it plans to enter the global edge cloud market through partnerships with global enterprises.

At this exhibition, Lablup is demonstrating 'GenAI Desktop,' an easy-to-use AI application integration platform for the general public. This platform is a tool that helps users easily understand and utilize AI. In addition, through Backend.AI FastTrack, attendees can directly demo the chatbot solution 'Talkativot,' developed by Lablup through fine-tuning large language models to fit specific needs. FastTrack is a B2B solution aimed at corporations and research institutions, simplifying and automating AI development through an MLOps platform. This platform manages all processes from data analysis, sorting, and processing to AI training, evaluation, and service deployment rapidly and effectively.

Lablup's CEO, Jeongkyu Shin, stated, "Lablup aims to 'Make AI Scalable' this year and will leap to become a platform company leading the global AI market through edge cloud-based AI solutions and generative AI application platforms." He added, "Lablup, participating as a silver sponsor at the NVIDIA GTC 2024 to be held in March, will lead the latest AI trends not only in the domestic market but also in the global AI."

Lablup to Showcase AI Edge Cloud Solutions and Generative AI Application Platform at MWC 2024

 ![](https://cdn.lablup.com/Gen_AI_Desktop_bf679fb88c.png) 

AI research and development platform company Lablup announced that will participate in the world's largest consumer electronics and IT trade show, CES 2024, in Las Vegas, USA. on January 9.

At the exhibition, Lablup will unveil 'GenAI Desktop,' a generative AI application integration platform for general users, and will introduce a demonstration of using generative AI applications directly from anywhere. 'GenAI Desktop' is a platform that allows applications that utilize AI models to run like desktop apps anywhere, and is characterized by the fact that users can select and run the services they need like an app, without having to have detailed knowledge of AI operations. One of GenAI Desktop's native apps, 'Talkativot,' which allows users to fine-tune large language models that are the hottest theme around the world and use them as chatbots. It will be demonstrated at Lablup's booth (#63301-51).

Lablup will also demonstrate Backend.AI FastTrack, a MLOps platform that simplifies and automates generative AI development as a B2B solution for enterprises and research labs. FastTrack quickly and effectively manages every step of the pipeline, from analyzing, classifying, and processing the data that makes up an AI project to training, evaluating, and deploying services. It makes all AI tasks replicable and sharable, so repeated processes can be easily reused or shared. Backend.AI FastTrack, evolved for the era of generative AI, is also available at the booth for running AI services at scale on GPU hyperscaler clouds.

Lablup, which is participating in CES for the second consecutive year, has been invited to participate in the K-Tech Forum@CES 2024 Korea pavilion, organized by KOTRA, as a startup panelist in the field of AI, and will share its experience in collaborating with global companies such as NVIDIA, PureStorage, NetApp, and Dell as technology partners and exporting to the UK and Brazil.

"Lablup has been constantly thinking and innovating on different ways to make AI easy, safe, and convenient for the world to use," said Jeongkyu Shin, CEO of Lablup. "This CES will be the beginning of introducing our solution to the world wide public on how to use AI in a friendly way." He added.

Lablup Unveils GenAI Application Integration Platform "GenAI Desktop" at CES

![](https://cdn.lablup.com/ai_relib_award_9d1147b73b.JPG)

The 'AI Reliability and Quality Award' is the first award system in Korea that focuses on AI reliability and quality, and aims to spread awareness of AI reliability and support the promotion of excellent domestic AI products and services by selecting and awarding AI products and services (including data) that have secured reliability.

Backend.AI is an all-in-one platform that integrates the entire process of researching and developing AI and creating services, and is an enterprise solution that specializes in the fields of super large-scale AI, high-performance computing, and supercomputing. It is installed on on-premises servers and cloud environments, and is responsible for smooth distributed execution and management of deep learning model design and training processes. Based on Linux container and VM technology, it securely distributes and executes AI and high-performance computing code on various platforms, and automates complex AI development processes that mobilize thousands of computers with GUI and MLOps functions without complex settings.

Backend.AI integrates the latest storage and AI acceleration hardware technologies on top of its proprietary orchestrator and scheduler, along with its own patented GPU partitioned virtualization technology that significantly reduces the cost of GPU usage. This enables the company to achieve industry-leading performance, efficiently reducing the cost of massive and costly AI training and the rapidly growing number of AI services.

Backend.AI is open-sourced so that anyone can develop AI without barriers, and Backend.AI Enterprise, an enterprise version with many features for enterprises and institutions, is recognized as an AI platform that integrates high performance and convenience, and is used by more than 70 enterprises and institutions, including Korean conglomerates such as Samsung Electronics, KT, and LG Electronics, as well as the Bank of Korea, the Health Insurance Review and Assessment Service, and GIST. It was also selected as the only DGX-Ready software in the Asia-Pacific region that is optimized to operate NVIDIA's AI supercomputer, DGX, and is currently exporting Backend.AI Enterprise to Latin America and Europe.

Founded in 2015, Lablup secured a Series A investment of 10.5 billion won earlier this year. "Backend.AI technically solves various reliability problems that occur in the process of AI development and operation.

"Backend.AI is helping many companies to create highly reliable and high-quality AI. We are pleased that Backend.AI itself has been recognized for achieving the reliability and quality requirements of various AI developments." "This award will be a great opportunity for Backend.AI to secure customers and users, and will help us to be more active in the global AI market." said Shin.

Backend.AI Receives AI Reliability and Quality Validation to Demonstrate Excellence

![Backend.AI package.jpg](https://cdn.lablup.com/Backend_AI_package_1276cda2a8.jpg)

Lablup announced today that will participate in "SC23 (November 12-17)" at the Colorado Convention Center (Denver) in the United States. The conference is an event for high-performance computing, networking, and storage analytics, and attracts top domestic and international companies such as AMD, ARM, AWS, IBM, Intel, NVIDIA, and SK hynix.

Lablup will introduce its platform, Backend.AI, for constructing a super-giant AI ecosystem at this year's SC23. Not only does Backend.AI support a variety of GPUs, including NVIDIA CUDA and AMD ROCm, it also offers optimized features for various NPUs and AI semiconductors. It supports Google TPU, Graphcore, Rebellion Atom, and Furiosa Warboy, among others, along with optimization features. Utilizing a GPU-NPU integrated pipeline to train AI models on GPUs and serve them on AI semiconductors, it achieves both the highest performance and the lowest cost.

Furthermore, Backend.AI is the first globally to implement GPU Direct Storage technology in a container cluster, providing tremendous data I/O speeds by directly connecting the GPU and network storage. This has led to partnerships with the top three in the high-speed storage market: PureStorage, NetApp, and Dell EMC. In addition, Backend.AI was the first and only one selected for the NVIDIA DGX-Ready program in the Asia-Pacific region and was chosen as a partner of the NVIDIA AI Accelerated Program along with VMWare, RedHat, and Canonical.

At SC23, Backend.AI will be demonstrating LLM services that can be directly developed and operated using this technology. It plans to announce that it can develop and serve LLM services in an on-premises environment through Backend.AI's automation system, FastTrack, by fine-tuning a variety of large public language models, including the Llama2 model disclosed by Meta, for various application areas. They will also showcase VisuTale, a multimodal demo that creates stories and generates images based on user-provided photos, subsequently creating a picture book.

"We are excited to introduce Backend.AI, which is optimized for high-performance computing, along with various supercomputer solutions, to supercomputing researchers around the world. We will continue to strive to achieve performance and cost innovations and overwhelming convenience through Backend.AI in technology-leading fields that require huge amounts of computational resources," said Jeongkyu Shin, CEO of Lablup.

Lablup introduces the hyperscaled AI service built from Backend.AI at SC23

![2023-08-llm-chatbot-screenshot.png](https://cdn.lablup.com/2023_08_llm_chatbot_screenshot_e548d4462f.png)

Lablup wowed developers at PyCon Korea 2023, held August 11-13, with features and demos that automate the process of creating, fine-tuning, and servicing large language models with Backend.AI, its AI R&D and services platform software.

In the demo, various open-source large language models, including the recently released Llama2 model from Meta, were fine-tuned for Korean using Backend.AI's automation system, FastTrack. The chatbot, which was created on the fly with the Llama2 model fine-tuned in 12 hours, was serviced during the event via Lablup's NVIDIA H100 cloud farm with a dedicated UI, attracting the attention of many participants.

Backend.AI provides leading features in various AI development and service fields based on ▲ maximizing GPU cluster utilization, ▲ intuitive user experience and management, ▲ low-cost/low-latency AI model and service technology, and ▲ support for various AI semiconductors. With the unveiling of its large-scale language model development, fine-tuning, and service automation capabilities, Lablup has demonstrated that Backend.AI can be utilized for full-scale generative AI applications such as task automation, research, and programming code development.

"We have dramatically simplified the process of developing and integrating various services based on large language models into enterprises, institutions, research institutes, and organizations that require security. We are confident that any organization can use Backend.AI to easily fine-tune and operate their own unique large language model to meet their needs."

Currently in beta testing with select Backend.AI customers, the big language model integration will be available to on-premises customers as part of the general availability of Backend.AI in September. The service will also be on display at KSC2023 (August 21-23, The K Hotel Seoul), Korea's largest supercomputing conference and exhibition.

> Automatically translated from Korean Article

Lablup to Debut Massive Language Model Integration Development and Services at PyCon Korea 2023

Lablup Inc. (CEO Jeongkyu Shin), an AI research, development, and service platform company, announced that it has finalized a **Series A investment** of **KRW 10.5 billion**. LB Investment, IBK, K2 Investment Partners, and Daesung Startup Investment participated in the round.

With the recent ChatGPT craze increasing interest in generative AI services, investors recognized the value of Lablup, which develops and sells Backend.AI, a platform that enables companies to efficiently operate GPUs, AI semiconductors, and data required for AI training and services.

Currently, Backend.AI is used in a variety of AI development and service fields, including the development of large language models and generative AI for pictures, text, music, etc. and the use of AI education platforms, based on maximizing GPU cluster utilization, intuitive user experience and management, low-cost/low-latency AI model and service technology, and support for various AI semiconductors.

Lablup's Backend.AI is used by large Korean companies such as Samsung Electronics, KT, LG Electronics, CJ, and more than 70 other companies and organizations, including the Bank of Korea, the Korea Health Insurance Agency, and GIST, and has been selected as NVIDIA's DGX-Ready software, the only total AIOps platform solution for AI development and services in the Asia Pacific region.

"As interest in AI R&D and services continues to grow globally, so does the demand for solutions that can provide much-needed resources for GPU utilization and cost savings," said Jeongkyu Shin, CEO of Lablup. "This investment will enable us to play a leading role in the AI platform solution market, both domestically and globally."

> Automatically translated from Korean Article

Lablup, an AI research, development, and services platform, raises 10.5B KRW (Approx. 80K USD) in Series A funding

Lablup Inc. (CEO Jeongkyu Shin) today announced that it will introduce Backend.AI R3 at the Mobile World Congress (MWC) 2023 in Barcelona, Spain.

![2023_02_mwc_1.jpg](https://cdn.lablup.com/2023_02_mwc_1_2380bd8e59.jpg)

The company will introduce Backend.AI, a platform that enables companies to efficiently operate GPUs, AI semiconductors, and data for AI training and services, to the European and global telecommunications industry and mobile-related companies at the show.

Backend.AI is a total AIOps platform solution for artificial intelligence (AI) development and services that is the only DGX-Ready software from NVIDIA in the Asia-Pacific region. It enables efficient use of the compute and power required to train very large AI models. The Backend.AI platform automates the AI research and service development process and provides the necessary resources within an organization to enable anyone to create and service AI without time and cost constraints. Recently, Backend.AI's MLOps platform, FastTrack, and Forklift services were integrated into an enterprise solution. This enables faster AI service development cycles, reducing the cost and time of AI development for various organizations.

"MWC 2023 is a place to see the reality of digital transformation that is revolutionizing existing industries," said Jeongkyu Shin, CEO of Lablup. "New technologies and business models are emerging through the convergence of innovation networks and AI." "Lablup is preparing AI model services based on Backend.AI that respond to various needs, from domestic telecommunications companies and giant language models to generation models, and MWC 2023 will be the place to introduce how global companies that want to adopt AI can achieve convenience and cost flexibility together."

> Automatically translated from Korean Article

Lablup to Participate in MWC 2023, Introduce AI R&D Platform to Increase Cost Competitiveness of Ultra-Large AI Model Services Like ChatGPT

*   Providing the best technology for container-based AI/ML model development and operational efficiency
*   Expected synergies between the two companies...from implementing GPU environment for AI/HPC to continuous operation management

Hyosung Information Systems (CEO Yang Jeong-kyu), a data infrastructure and solutions company, and Lablup Inc. (CEO JeongKyu Shin), a company specializing in hyper-scale AI development platforms, announced on the XXth that they have signed a memorandum of understanding (MOU) to expand the domestic AI (artificial intelligence), ML (machine learning), and HPC (high-performance computing) markets and strengthen collaboration, including joint sales and marketing.

Successful AI projects require advanced strategies for efficient utilization of GPU computing resources, effective storage and performance maintenance of growing data, easy development of AI models, and continuous operation management.

Hyosung Information Systems is strengthening its pursuit of the GPU-based AI platform business with Lablup, which is recognized globally for its AI development platform and GPU partitioned virtualization technology. Based on proven products and various project know-how, the two companies will provide the best infrastructure environment for customers' AI projects, from GPU environment implementation to operation management.

Hyosung Information Systems simplifies complex AI tasks for companies through its 'AI Integrated Platform' that combines performance-optimized GPU server infrastructure and AI model operation management system. It provides integrated proposals from GPU-based data computation to large data storage and automated management and operation using storage optimized for high-performance computing, actively supporting customers' data-based digital transformation.

Lablup provides Backend.AI, an all-in-one enterprise cluster platform for AI development and services. Backend.AI provides all the functions required in the era of AI development and service, including self-developed container-based GPU partitioned virtualization technology, deep learning model distributed training and ultra-large model development platform technology, and GPU-based ultra-low cost/ultra-low latency AI model service technology. It is currently being used by leading domestic and international companies and organizations, including Samsung Electronics, KT, LG CNS, SK, Lotte Communications, CJ, and others.

"Based on our technology partnerships such as NVIDIA's AI Accelerated Program and DGX-Ready Software Program, as well as our proven technology through large-scale enterprise and institutional customers at home and abroad, we will enhance our synergies for our customers' successful AI projects." "From implementing a GPU infrastructure environment to developing effective AI models based on containers and optimizing service management, our customers will be able to accelerate the introduction of AI services and create business results based on AIOps," said Jeongkyu Shin, CEO of Lablup.

"Based on the data and AI-related technologies that both companies have accumulated in Korea, we will provide the best infrastructure environment for customers who are experiencing difficulties with GPU virtualization and resource management," said Yang Jeong-kyu, CEO of Hyosung Information Systems.

> About Lablup Inc. ([https://www.lablup.com](https://www.lablup.com))
> 
> Founded in 2015,  a company specializing in hyper-scale AI development platforms, has been recognized globally for its AI development and service platform Backend.AI, which is the first in the Asia-Pacific region to be certified as NVIDIA DGX-Ready Software.
> 
> Backend.AI ([www.backend.ai](https://www.backend.ai)) enables effective management of AI/ML/HPC from R&D to business services to AI service inference and delivery through one consistent platform and is utilized by various customers such as Samsung Electronics, KT, LG CNS, SK, Lotte Communications, SK, Bank of Korea, and ETRI.
> 
> > Automatically translated from Korean

Hyosung Information Systems and Lablup Strengthen Market Penetration of GPU-based (Hyper-scale) AI Platforms

During the 2022 GTC keynote, NVIDIA introduced the AI Accelerated program partners. Among the more than 100 companies was Lablup, a Korean company that developed Backend.AI, a hyperscale AI R&D and service platform.

NVIDIA's AI Accelerated program is for AI-enabled application developers and solution partners who leverage the NVIDIA AI platform to deliver faster, more efficient, and better results for their customers. The program includes some of the world's best-known companies, including Adobe, VMware, and Canonical.

Having already been certified as the first and only DGX-ready software from NVIDIA in the Asia-Pacific region last summer, Backend.AI's inclusion in the AI Accelerated program is a recognition of its global competitiveness and excellence in the MLOps and AI services market.

Backend.AI is a total platform solution for cloud and on-premises that builds an AIOps environment for AI development and service, automating the entire process from AI R&D to service through a standardized platform and providing the necessary resources through an already owned AI cluster or cloud, enabling anyone to create and service AI without time and cost constraints.

Users can conveniently develop AI models on laptops, editors, and professional terminals, and developers can integrate Backend.AI into their own software and platforms based on a powerful SDK. In addition, operators can intuitively perform individual, group, and domain-specific management through the control panel and view various real-time statistics through the dashboard.

"Backend.AI's selection as a partner in NVIDIA's AI Accelerated program, a leader in AI technology, will enable the MLOps ecosystem to further accelerate and lead the AI development process for all users at home and abroad through more advanced hyperscale AIOps," said Jeongkyu Shin, CEO of Lablup.

> Automatically translated from Korean Article

Lablup Announces Backend.AI as a Partner in NVIDIA's AI Accelerator Program

Today announced that Lablup Inc. has been selected as a global technology partner in the Technolpgy Alliance Partners (TAP) program with Pure Storage, a leading provider of all-flash storage platforms.

The TAP program recognizes Technolpgy Alliance Partners who collaborate with Pure Storage to build and deploy pre-tested joint solutions, including AI, analytics, cloud, and modern data protection, to deliver and build better businesses.

Lablup's Backend.AI is a solution that efficiently leverages hardware, software, and multi-cloud-sourced cloud and data center compute resources to run and manage large-scale computational workloads, including large-scale AI model development and high-performance computing, by optimizing the performance of accelerated computing, including GPUs on Pure Storage FlashBlade, to efficiently manage compute resources to maximize GPU usage without changing the user experience.

(Reference: https://www.purestorage.com/partners/technology-alliance-partners/lablup.html)

"The development of AI research is no longer just for research, but for real-world applications, and the infrastructure for such development is becoming large-scale and sophisticated," said Jeongmook Kim, Chief Operating Officer of Lablup. "By becoming a TAP with Pure Storage, we are leading the global market in providing highly usable hardware to build such infrastructure and support for optimal solutions that maximize that hardware to further accelerate the AI development process and optimize everything from deployment, usability, and utilization of AI infrastructure."

Lablup Partners with Pure Storage to Expand Global Footprint

Today Lablup Inc. announced that Backend.AI, a solution that efficiently leverages cloud and data center compute resources to run and manage large-scale computational workloads, including large-scale artificial intelligence (AI) model development and high-performance computing, is the first in Asia Pacific to be certified as NVIDIA DGX-Ready Software.

NVIDIA DGX-Ready Software is a validated, enterprise-grade software solution that is fully tested and certified for use on clusters of NVIDIA DGX systems, and Backend.AI is the first company in Asia Pacific to be certified.

Backend.AI optimizes the performance of NVIDIA DGX systems and accelerated computing, including NVIDIA GPUs, to efficiently manage compute resources to maximize GPU utilization without changing the user experience. In addition, Backend.AI has its own patented container-based GPU partitioned virtualization technology, which allows users to partition expensive GPUs into multiple GPUs as needed to maximize GPU usage.   
(Reference: [https://www.backend.ai/partners/nvidia)](https://www.backend.ai/partners/nvidia))

"Backend.AI has been successfully supporting all types of AI platform customers, from educational institutions to Hyperscale AI Infra," said Jeongkyu Shin, CEO of Lablup Inc. We are delighted to be the first in Asia-Pacific to be certified as NVIDIA DGX-ready software, and we look forward to helping our users, both in Korea and abroad, not only accelerate their AI development process, but also optimize their deployment, usability, and utilization of their AI infrastructure."

NVIDIA DGX systems accelerate large-scale AI initiatives to help organizations solve key challenges in healthcare, manufacturing, commerce, science, and more. The NVIDIA DGX-ready software program helps customers succeed by validating enterprise software solutions like Lablup's Backend.AI that streamline workflows and optimize utilization across AI clusters.

> Automatically translated from Korean

lablup.com

lablup.com

Press Resources

Press Releases