Ten
- Company typePrivate
- Founded2020
- HeadquartersSeoul, South Korea
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
Ten firmographics
Firmographics- Name
- Ten
- Legal name
- 주식회사 텐
- Website
- https://ten1010.io
- Company type
- Private
- Founded year
- 2020
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Ownership category
- akta.pro rank
Ten industry classification
Industry- Product category
- AI Infrastructure MLOps
- NAICS
- Software Publishers (513210), Computer Systems Design and Related Services (54151)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370)
- akta.pro primary industry
- AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers) (HDAAAAAG)
- akta.pro secondary industries
- AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ), Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem) (HDAEANAC), Model Development & Training Platforms (AutoML, Notebooks, Feature Stores) (HDAEANAB), AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety) (HDAEANAF), AI/ML Solution Integration & MLOps Enablement (BPAEAAAJ)
Keywords
Where Ten is headquartered
LocationHeadquarters
- HQ city
- Seoul
- HQ country
- South Korea
- HQ region
- Asia
Offices1 record
Markets served
Ten business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Marketing or Sales, Operations, Infrastructure
Revenue model
- AI Pub Dev (Subscription/Managed Service): GPU 자원 운용 기술 기반 AI 개발 플랫폼. 완전 관리형 서비스로 모델 학습, 리소스 및 워크로드 관리. 팀/프로젝트 단위로 GPU 인프라 리소스 할당 및 모니터링.
- AI Pub Ops (Subscription/Managed Service): AI 모델 서빙 및 운영을 위한 통합 MLOps 플랫폼. Web UI 기반 서비스 생성, 모니터링, 관리. 완전 관리형 서비스로 인프라 운영 대행 옵션 포함.
- RA:X Infrastructure Design Consulting: AI 인프라 설계 컨설팅 서비스. 고객의 모델과 데이터 샘플을 TEN의 레퍼런스 아키텍처에서 직접 실행하여 최적 GPU 구성을 제안하는 유료 맞춤 컨설팅. 벤치마킹 기반 데이터 의사결정 지원.
- TEN MDC (Modular Data Center): 모듈형 데이터센터 구축 및 운영. GPU 세대 변화 대응 유연 구조, 고밀도 AI 서버 수용. 전산실, UPS실, 배터리실, 냉각 설비 모듈 단위 설계 및 조립.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Subscription | Annual | Custom enterprise pricing — not publicly disclosed |
Go-to-market motion2 records
Distribution channels3 records
Marketing channels7 records
Ten product offering
Product offeringCore offering
TEN sells an AI infrastructure MLOps platform composed of four integrated products: RA:X (intelligent infrastructure design and benchmarking for AI workloads), Coaster (a Kubernetes-extended container platform enabling GPU spatial partitioning into up to 100 blocks per GPU), AI Pub Dev (a managed platform for AI model training and development with GPU resource scheduling, MIG-based partitioning, and real-time monitoring), and AI Pub Ops (an integrated MLOps platform for AI model serving, non-disruptive rolling updates, monitoring, and multi-tenant operations). Offerings are sold to enterprises, public institutions, national universities, and research institutes in Korea through direct enterprise sales and government procurement channels.
Product overview
Ten (주식회사 텐) is an AI infrastructure MLOps company offering four integrated products: RA:X (intelligent AI infrastructure design and benchmarking), Coaster (Kubernetes-extended GPU container platform enabling GPU partitioning), AI Pub Dev (AI development/training resource management), and AI Pub Ops (AI service serving and operations). The portfolio forms a unified AI infrastructure orchestration system — RA:X designs and validates the optimal hardware architecture, Coaster provides the GPU-partitioned container substrate, and AI Pub Dev and AI Pub Ops sit on top to manage GPU resource allocation, scheduling, and monitoring for AI development and production serving respectively.
Differentiator
Problem solved
Functional benefit
Brands
- AI Pub: Web UI-based MLOps platform for AI development and operations, built on Coaster, supporting GPU resource management, workflow scheduling, and monitoring.
- AI Pub Dev
- AI Pub Ops
- Coaster
- RAX
Products and services
- RA:X Intelligent AI infrastructure design service that benchmarks customer AI models and workloads against reference GPU architectures (DGX H100, DGX A100, L40S, L40, A40, T4) and recommends optimal GPU, CPU, memory, storage, and network configurations. Real-environment validated performance is delivered with monitoring of GPU utilization, memory bandwidth, network latency, and storage.
- Coaster Kubernetes-extended container platform that splits a single GPU's compute and memory into up to 100 blocks for stable multi-tenant use, with full Kubernetes API support, Resource Group Controller for user-group resource sharing, scheduler queue management, and priority changes. The core extension component resource-group-controller is published as open source on GitHub.
- AI Pub Dev GPU resource management tool for AI development and training built on the Coaster container platform. Supports multi-user GPU resource sharing, MIG partitioning (1-7 instances per GPU), workload management with scheduler queue control, resource usage tracking, and real-time cluster-, node-, and project-level GPU monitoring via Web UI.
- AI Pub Ops Integrated MLOps platform for AI model serving and operations built on Coaster's GPU 100-block partitioning. Provides Web UI-based service creation and management, non-disruptive rolling updates, real-time monitoring, service-level alerting, and GPU block-level allocation. Includes NetApp Astra integration for microservice data management, backup, migration, and rollback, plus NeuVector-based container security scanning. Can reduce infrastructure costs by up to 90% compared to full-GPU allocation.
- TEN MDC (Modular Data Center) Modular data center construction and operations offering from TEN, including prefabricated electrical rooms, UPS rooms, battery rooms, and cooling system modules designed for high-density AI server deployment and rapid adaptation to GPU generation changes.
Quantifiable outcome
- Up to 90% infrastructure cost reduction by using only 1/10 of server resources through GPU block splitting
- +5 more outcomes
Companies that use Ten
Customer profileNamed customers5 records
Segments4 records
Ideal customer profiles4 records
Ten technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- No
- API docs
- API detail
Core technology
AI maturity
App detail
Integration3 records
AI capability4 records
Feature9 records
Ten partnerships and signals
Strategic signalPartnerships
One partnership is on record.
- NVIDIAcoreTEN's platform is deeply integrated with NVIDIA GPU ecosystem. Supports NVIDIA MIG (Multi-Instance GPU) technology for hardware-level GPU partitioning (up to 7 instances on A100/H100). Coaster and AI Pub are positioned within the AI technology stack alongside NVIDIA's AI ecosystem. Blog extensively covers NVIDIA GTC 2026, Blackwell architecture, NVLink, and NVIDIA's GPU scheduling tools donated to CNCF.
Scale indicators5 records
Recent moves6 records
Expansion highlights7 records
Ten competitors and assessment
Company assessmentDirect peers
- Run:ai: GPU virtualization and orchestration platform built natively on Kubernetes. Most direct global competitor to TEN's Coaster and AI Pub, offering dynamic MIG partitioning, fractional GPU sharing, and multi-tenant scheduling. Acquired by NVIDIA in 2024.
- CoreWeave: Large-scale GPU cloud built on Kubernetes with custom scheduling and bare-metal GPU partitioning. Comparable to TEN's end-to-end AI infrastructure thesis (hardware + software + managed services), though operating at much larger scale and globally.
- Lambda Labs: GPU cloud and AI infrastructure provider serving enterprise AI teams, with reference architectures for DGX-class hardware and Kubernetes-based cluster management. Directly comparable to TEN's RA:X consulting plus AI Pub platform offering.
- Rafay Systems: Kubernetes management and orchestration platform with multi-cluster, multi-cloud operations and GPU workload support. Comparable to TEN's Coaster multi-cluster management and AI Pub Ops operational layer.
- Paperspace (DigitalOcean): GPU cloud for AI/ML training and inference, now part of DigitalOcean. Comparable to TEN's AI Pub Dev/Ops value prop of easy-to-use, managed GPU infrastructure for AI developers, with self-service and Kubernetes integration.
- Vast.ai: GPU marketplace and cloud platform enabling fractional GPU access and multi-tenant scheduling. Comparable to TEN's spatial partitioning economics, though primarily consumer/developer oriented rather than enterprise.
Emerging players
- Spectro Cloud: Kubernetes management platform with a focus on AI/ML workloads, GPU support, and enterprise-grade cluster operations. Comparable in target buyer (enterprise platform teams) and Kubernetes-native approach to Coaster.
- Anyscale: AI compute platform built on Ray for distributed training, serving, and orchestration. Comparable to TEN's AI Pub Ops for production AI workloads and Coaster's resource scheduling, with broader enterprise MLOps overlap.
Broad incumbents
- HPE Determined AI: Enterprise ML training platform now part of HPE, with deep Kubernetes and Slurm integration for distributed training. Comparable to TEN's AI Pub Dev as a managed AI training environment for enterprise teams.
Others
- NVIDIA GPU Operator: NVIDIA's open-source Kubernetes operator for provisioning GPU drivers, runtime, and monitoring. Adjacent enabling layer that TEN builds on top of and partially competes with via Coaster's extended GPU partitioning features.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
Ten social profiles
Digital presenceTen financial estimates
Financial estimateRevenue estimate
Valuation estimate
Ten leadership team
Management profileNumber of profiles
Profiles1 record
Ten funding detail
Funding detailFunding overview
Funding rounds3 records
Investors5 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Ten M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Ten
What does Ten do?
TEN sells an AI infrastructure MLOps platform composed of four integrated products: RA:X (intelligent infrastructure design and benchmarking for AI workloads), Coaster (a Kubernetes-extended container platform enabling GPU spatial partitioning into up to 100 blocks per GPU), AI Pub Dev (a managed platform for AI model training and development with GPU resource scheduling, MIG-based partitioning, and real-time monitoring), and AI Pub Ops (an integrated MLOps platform for AI model serving, non-disruptive rolling updates, monitoring, and multi-tenant operations). Offerings are sold to enterprises, public institutions, national universities, and research institutes in Korea through direct enterprise sales and government procurement channels.
Is Ten a public or private company?
Ten is a private company. It is classified as venture growth investor backed and is currently operating.
When was Ten founded?
Ten was founded in 2020. It employs 11 to 50 people.
Where is Ten based?
Ten is headquartered in Seoul, South Korea, in the Asia region.
How does Ten make money?
Four revenue lines are on record. AI Pub Dev (Subscription/Managed Service) is the primary driver. The others are AI Pub Ops (Subscription/Managed Service), RA:X Infrastructure Design Consulting and TEN MDC (Modular Data Center).
Who are Ten's main competitors?
Direct peers on record are Run:ai, CoreWeave, Lambda Labs, Rafay Systems, Paperspace (DigitalOcean) and Vast.ai. Emerging players are Spectro Cloud and Anyscale. HPE Determined AI is listed as a broad incumbent. NVIDIA GPU Operator is listed as an others.
Does Ten have an API?
No public API is recorded for Ten.
What industry is Ten in?
Ten's product category is AI Infrastructure MLOps. Its primary akta.pro industry code is HDAAAAAG, AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers), with a secondary code of HDAEANAJ, AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs). Its NAICS code is 513210 and its SIC code is 7372.