ENG
KORWe provide an optimized HPC cloud environmentfor AI·DX and engineering simulation.
Accelerate your simulation projects with all the benefits of Virtual Labs
-
Expert Engineers
Experts who understand complex engineering simulations design and support your environment directly.
-
Pay-per-use
A flexible and rational cost structure that bills only what you use.
-
High Performance
High-performance CPU·GPU and optimized networks handle large computations fast and reliably.
-
Security
Modern encryption and access control keep industrial secrets and data safe.
-
Flexibility
Maintain optimal performance and cost efficiency as projects evolve.
-
Ready-to-use
Instantly available environments let you start simulations right away.
Charge as much time as you want!
Longer usage, smarter pricing.
- Workstation
- Cluster
vCPUs
Runtime
-
Computing Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory16Credit / Time 4,000BenchmarkTestSingle-Core Score2036Multi-Core Score7870GPU ScorevCPUs16Memory32Credit / Time 8,000BenchmarkTestSingle-Core Score2079Multi-Core Score12191GPU ScorevCPUs32Memory64Credit / Time 16,000BenchmarkTestSingle-Core Score2084Multi-Core Score16741GPU ScorevCPUs48Memory96Credit / Time 24,000BenchmarkTestSingle-Core Score2074Multi-Core Score17515GPU ScorevCPUs64Memory128Credit / Time 32,000BenchmarkTestSingle-Core Score2111Multi-Core Score19545GPU ScorevCPUs96Memory192Credit / Time 48,000BenchmarkTestSingle-Core Score2086Multi-Core Score19028GPU ScorevCPUs192Memory384Credit / Time 95,900BenchmarkTestSingle-Core Score2055Multi-Core Score18018GPU Score
-
HPC Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory32Credit / Time 5,000BenchmarkTestSingle-Core Score2088Multi-Core Score8054GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs16Memory64Credit / Time 10,000BenchmarkTestSingle-Core Score2047Multi-Core Score12445GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs32Memory128Credit / Time 20,000BenchmarkTestSingle-Core Score2109Multi-Core Score16786GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs48Memory192Credit / Time 30,000BenchmarkTestSingle-Core Score2099Multi-Core Score17694GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs64Memory256Credit / Time 40,000BenchmarkTestSingle-Core Score2071Multi-Core Score21286GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs96Memory384Credit / Time 59,900BenchmarkTestSingle-Core Score2193Multi-Core Score20460GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs192Memory786Credit / Time 119,800BenchmarkTestSingle-Core Score2158Multi-Core Score19550GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.
-
HPC(Memory) Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory64Credit / Time 5,400BenchmarkTestSingle-Core Score2176Multi-Core Score8308GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs16Memory128Credit / Time 10,800BenchmarkTestSingle-Core Score2141Multi-Core Score12699GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs32Memory256Credit / Time 21,600BenchmarkTestSingle-Core Score2010Multi-Core Score15166GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs48Memory384Credit / Time 32,400BenchmarkTestSingle-Core Score2054Multi-Core Score17914GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs64Memory512Credit / Time 43,200BenchmarkTestSingle-Core Score2118Multi-Core Score19182GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs96Memory768Credit / Time 64,800BenchmarkTestSingle-Core Score2131Multi-Core Score20127GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs192Memory1536Credit / Time 129,500BenchmarkTestSingle-Core Score2102Multi-Core Score18298GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.
-
GPU(T4) Intel Xeon Platinum 8259CL (Cascade Lake) 2.5GHz | NVIDIA T4 16GB * 1 GPUvCPUs8Memory32Credit / Time 4,700BenchmarkTestSingle-Core Score1145Multi-Core Score4143GPU Score78370NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs16Memory64Credit / Time 8,100BenchmarkTestSingle-Core Score1126Multi-Core Score6774GPU Score80206NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs32Memory128Credit / Time 15,000BenchmarkTestSingle-Core Score1151Multi-Core Score9758GPU Score82229NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs48Memory192Credit / Time 25,400BenchmarkTestSingle-Core Score1164Multi-Core Score10514GPU Score77148NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs64Memory256Credit / Time 30,000BenchmarkTestSingle-Core Score1104Multi-Core Score11290GPU Score83826NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.
-
GPU(A10G) AMD EPYC Milan 3.3GHz | NVIDIA A10G 22.4GB * 1 GPUvCPUs8Memory32Credit / Time 5,900BenchmarkTestSingle-Core Score1376Multi-Core Score5301GPU Score162804NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs16Memory64Credit / Time 8,700BenchmarkTestSingle-Core Score1335Multi-Core Score7813GPU Score162464NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs32Memory128Credit / Time 14,200BenchmarkTestSingle-Core Score1368Multi-Core Score11797GPU Score162864NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs48Memory192Credit / Time 29,100BenchmarkTestSingle-Core Score1324Multi-Core Score13002GPU Score165557NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs64Memory256Credit / Time 25,300BenchmarkTestSingle-Core Score1326Multi-Core Score13940GPU Score167952NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs96Memory384Credit / Time 45,600BenchmarkTestSingle-Core Score1329Multi-Core Score13781GPU Score168749NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs192Memory768Credit / Time 91,200BenchmarkTestSingle-Core Score1307Multi-Core Score14102GPU Score167478NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.
-
GPU(L4) AMD EPYC 7R13 3.4GHz | NVIDIA L4 22.4GB * 1 GPUvCPUs8Memory32Credit / Time 5,400BenchmarkTestSingle-Core Score1744Multi-Core Score6700GPU Score153070NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs16Memory64Credit / Time 8,100BenchmarkTestSingle-Core Score1741Multi-Core Score10077GPU Score155509NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs32Memory128Credit / Time 13,500BenchmarkTestSingle-Core Score1738Multi-Core Score14166GPU Score155681NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs48Memory192Credit / Time 26,800BenchmarkTestSingle-Core Score1754Multi-Core Score16892GPU Score155387NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs64Memory256Credit / Time 24,300BenchmarkTestSingle-Core Score1736Multi-Core Score16623GPU Score155986NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs96Memory384Credit / Time 43,000BenchmarkTestSingle-Core Score1680Multi-Core Score17598GPU Score153684NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs192Memory768Credit / Time 85,900BenchmarkTestSingle-Core Score1686Multi-Core Score16580GPU Score154847NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.
-
GPU(L40S) AMD EPYC 7R13 3.4GHz | NVIDIA L40S 44.7GB * 1 GPUvCPUs8Memory64Credit / Time 10,100BenchmarkTestSingle-Core Score1673Multi-Core Score6494GPU Score192690NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs16Memory128Credit / Time 14,400BenchmarkTestSingle-Core Score1694Multi-Core Score9945GPU Score198570NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs32Memory256Credit / Time 22,800BenchmarkTestSingle-Core Score1681Multi-Core Score14031GPU Score199569NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs48Memory384Credit / Time 48,800BenchmarkTestSingle-Core Score1699Multi-Core Score15912GPU Score199684NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs64Memory512Credit / Time 39,600BenchmarkTestSingle-Core Score1693Multi-Core Score16621GPU Score199675NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs96Memory768Credit / Time 74,100BenchmarkTestSingle-Core Score1677Multi-Core Score17756GPU Score199976NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs192Memory1536Credit / Time 148,200BenchmarkTestSingle-Core Score1685Multi-Core Score16100GPU Score199985NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.
-
AI(V100) Intel Xeon E5‑2686 v4 (Broadwell) 2.7GHz | NVIDIA V100 16GB * 1 GPU | Only LinuxvCPUs8Memory61Credit / Time 16,200BenchmarkTestSingle-Core Score748Multi-Core Score3284GPU Score172840NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.vCPUs32Memory244Credit / Time 64,500BenchmarkTestSingle-Core Score749Multi-Core Score7558GPU Score173874NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.vCPUs64Memory488Credit / Time 128,900BenchmarkTestSingle-Core Score751Multi-Core Score13598GPU Score175948NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.
-
AI(A100) Intel Xeon Platinum 8275L 3.0GHz | NVIDIA A100 40GB * 8 GPU | Only LinuxvCPUs96Memory1152Credit / Time 110,100BenchmarkTestSingle-Core Score1291Multi-Core Score15956GPU Score213187NVIDIA A100–based. Optimized for large-scale model training and inference. Supports FP64, TF32, and FP16 operations with NVLink and NVSwitch for efficient distributed training and multi-GPU scaling.
-
AI(H100) AMD EPYC 7R13 3.6GHz | NVIDIA H100 80GB * 1 GPU | Only LinuxvCPUs16Memory256Credit / Time 30,800BenchmarkTestSingle-Core Score1680Multi-Core Score6732GPU Score325474NVIDIA H100–based. Designed for large-scale LLMs and multi-node training. Maximizes training speed with FP8/TF32 precision and NVLink, delivering top-tier performance for massive fine-tuning and HPC workloads.vCPUs192Memory2048Credit / Time 272,300BenchmarkTestSingle-Core Score1709Multi-Core Score16009GPU Score334465NVIDIA H100–based. Designed for large-scale LLMs and multi-node training. Maximizes training speed with FP8/TF32 precision and NVLink, delivering top-tier performance for massive fine-tuning and HPC workloads.
-
AI(H200) Intel Xeon Scalable (Sapphire Rapids) 2.7GHz | NVIDIA H200 141GB * 8 GPU | Only LinuxvCPUs192Memory2048Credit / Time 312,200BenchmarkTestSingle-Core Score2048Multi-Core Score20489GPU Score384329NVIDIA H200–based. Next-generation GPU optimized for large-scale generative AI training and inference. Equipped with 141 GB of HBM3 memory and up to 4.8 TB/s bandwidth, supporting FP8, FP16, and BF16 operations for maximum efficiency in large-model workloads.
-
Computing Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory16Credit / Time 4,000BenchmarkTestSingle-Core Score2036Multi-Core Score7870GPU ScorevCPUs16Memory32Credit / Time 8,000BenchmarkTestSingle-Core Score2079Multi-Core Score12191GPU ScorevCPUs32Memory64Credit / Time 16,000BenchmarkTestSingle-Core Score2084Multi-Core Score16741GPU ScorevCPUs48Memory96Credit / Time 24,000BenchmarkTestSingle-Core Score2074Multi-Core Score17515GPU ScorevCPUs64Memory128Credit / Time 32,000BenchmarkTestSingle-Core Score2111Multi-Core Score19545GPU ScorevCPUs96Memory192Credit / Time 48,000BenchmarkTestSingle-Core Score2086Multi-Core Score19028GPU ScorevCPUs192Memory384Credit / Time 95,900BenchmarkTestSingle-Core Score2055Multi-Core Score18018GPU Score
-
HPC Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory32Credit / Time 5,000BenchmarkTestSingle-Core Score2088Multi-Core Score8054GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs16Memory64Credit / Time 10,000BenchmarkTestSingle-Core Score2047Multi-Core Score12445GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs32Memory128Credit / Time 20,000BenchmarkTestSingle-Core Score2109Multi-Core Score16786GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs48Memory192Credit / Time 30,000BenchmarkTestSingle-Core Score2099Multi-Core Score17694GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs64Memory256Credit / Time 40,000BenchmarkTestSingle-Core Score2071Multi-Core Score21286GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs96Memory384Credit / Time 59,900BenchmarkTestSingle-Core Score2193Multi-Core Score20460GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.vCPUs192Memory786Credit / Time 119,800BenchmarkTestSingle-Core Score2158Multi-Core Score19550GPU ScoreHigh-frequency, low-latency vCPUs optimized for workloads where single-thread performance is critical — such as CAD/EDA, financial computing, and real-time processing. Ideal for short-running jobs that demand maximum performance.
-
HPC(Memory) Intel Xeon Scalable (Sapphire Rapids) 3.2GHzvCPUs8Memory64Credit / Time 5,400BenchmarkTestSingle-Core Score2176Multi-Core Score8308GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs16Memory128Credit / Time 10,800BenchmarkTestSingle-Core Score2141Multi-Core Score12699GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs32Memory256Credit / Time 21,600BenchmarkTestSingle-Core Score2010Multi-Core Score15166GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs48Memory384Credit / Time 32,400BenchmarkTestSingle-Core Score2054Multi-Core Score17914GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs64Memory512Credit / Time 43,200BenchmarkTestSingle-Core Score2118Multi-Core Score19182GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs96Memory768Credit / Time 64,800BenchmarkTestSingle-Core Score2131Multi-Core Score20127GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.vCPUs192Memory1536Credit / Time 129,500BenchmarkTestSingle-Core Score2102Multi-Core Score18298GPU ScoreLarge-memory instances ideal for in-memory databases, large-scale data analytics and preprocessing, as well as CAD/CAE preprocessing. Designed to minimize swapping.
-
GPU(T4) Intel Xeon Platinum 8259CL (Cascade Lake) 2.5GHz | NVIDIA T4 16GB * 1 GPUvCPUs8Memory32Credit / Time 4,700BenchmarkTestSingle-Core Score1145Multi-Core Score4143GPU Score78370NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs16Memory64Credit / Time 8,100BenchmarkTestSingle-Core Score1126Multi-Core Score6774GPU Score80206NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs32Memory128Credit / Time 15,000BenchmarkTestSingle-Core Score1151Multi-Core Score9758GPU Score82229NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs48Memory192Credit / Time 25,400BenchmarkTestSingle-Core Score1164Multi-Core Score10514GPU Score77148NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.vCPUs64Memory256Credit / Time 30,000BenchmarkTestSingle-Core Score1104Multi-Core Score11290GPU Score83826NVIDIA T4–based. Well-suited for CAD, visualization, video editing, and general 3D training. Cost-effective for running multiple sessions.
-
GPU(A10G) AMD EPYC Milan 3.3GHz | NVIDIA A10G 22.4GB * 1 GPUvCPUs8Memory32Credit / Time 5,900BenchmarkTestSingle-Core Score1376Multi-Core Score5301GPU Score162804NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs16Memory64Credit / Time 8,700BenchmarkTestSingle-Core Score1335Multi-Core Score7813GPU Score162464NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs32Memory128Credit / Time 14,200BenchmarkTestSingle-Core Score1368Multi-Core Score11797GPU Score162864NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs48Memory192Credit / Time 29,100BenchmarkTestSingle-Core Score1324Multi-Core Score13002GPU Score165557NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs64Memory256Credit / Time 25,300BenchmarkTestSingle-Core Score1326Multi-Core Score13940GPU Score167952NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs96Memory384Credit / Time 45,600BenchmarkTestSingle-Core Score1329Multi-Core Score13781GPU Score168749NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.vCPUs192Memory768Credit / Time 91,200BenchmarkTestSingle-Core Score1307Multi-Core Score14102GPU Score167478NVIDIA A10G–based. Optimized for complex 3D rendering, high-resolution streaming, and AI inference/lightweight training.
-
GPU(L4) AMD EPYC 7R13 3.4GHz | NVIDIA L4 22.4GB * 1 GPUvCPUs8Memory32Credit / Time 5,400BenchmarkTestSingle-Core Score1744Multi-Core Score6700GPU Score153070NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs16Memory64Credit / Time 8,100BenchmarkTestSingle-Core Score1741Multi-Core Score10077GPU Score155509NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs32Memory128Credit / Time 13,500BenchmarkTestSingle-Core Score1738Multi-Core Score14166GPU Score155681NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs48Memory192Credit / Time 26,800BenchmarkTestSingle-Core Score1754Multi-Core Score16892GPU Score155387NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs64Memory256Credit / Time 24,300BenchmarkTestSingle-Core Score1736Multi-Core Score16623GPU Score155986NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs96Memory384Credit / Time 43,000BenchmarkTestSingle-Core Score1680Multi-Core Score17598GPU Score153684NVIDIA L4–based. Specialized for large-scale inference, computer vision, and video transcoding. Delivers low latency, high throughput, and power efficiency, making it ideal for production services and multi-session workloads.vCPUs192Memory768Credit / Time 85,900BenchmarkTestSingle-Core Score1686Multi-Core Score16580GPU Score154847NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.
-
GPU(L40S) AMD EPYC 7R13 3.4GHz | NVIDIA L40S 44.7GB * 1 GPUvCPUs8Memory64Credit / Time 10,100BenchmarkTestSingle-Core Score1673Multi-Core Score6494GPU Score192690NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs16Memory128Credit / Time 14,400BenchmarkTestSingle-Core Score1694Multi-Core Score9945GPU Score198570NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs32Memory256Credit / Time 22,800BenchmarkTestSingle-Core Score1681Multi-Core Score14031GPU Score199569NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs48Memory384Credit / Time 48,800BenchmarkTestSingle-Core Score1699Multi-Core Score15912GPU Score199684NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs64Memory512Credit / Time 39,600BenchmarkTestSingle-Core Score1693Multi-Core Score16621GPU Score199675NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs96Memory768Credit / Time 74,100BenchmarkTestSingle-Core Score1677Multi-Core Score17756GPU Score199976NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.vCPUs192Memory1536Credit / Time 148,200BenchmarkTestSingle-Core Score1685Multi-Core Score16100GPU Score199985NVIDIA L40S–based. Built for Gen-AI, high-performance inference, and RTX rendering. Large VRAM and latest Tensor/RT cores speed graphics+AI workloads—ideal for enterprise real-time, Omniverse, and high-density VDI.
-
AI(V100) Intel Xeon E5‑2686 v4 (Broadwell) 2.7GHz | NVIDIA V100 16GB * 1 GPU | Only LinuxvCPUs8Memory61Credit / Time 16,200BenchmarkTestSingle-Core Score748Multi-Core Score3284GPU Score172840NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.vCPUs32Memory244Credit / Time 64,500BenchmarkTestSingle-Core Score749Multi-Core Score7558GPU Score173874NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.vCPUs64Memory488Credit / Time 128,900BenchmarkTestSingle-Core Score751Multi-Core Score13598GPU Score175948NVIDIA V100–based. Well-suited for mid-scale model training and fine-tuning. Supports distributed training with FP16 Tensor Cores and NVLink, delivering stable performance at a balanced cost—ideal for education and PoC use cases.
-
AI(A100) Intel Xeon Platinum 8275L 3.0GHz | NVIDIA A100 40GB * 8 GPU | Only LinuxvCPUs96Memory1152Credit / Time 110,100BenchmarkTestSingle-Core Score1291Multi-Core Score15956GPU Score213187NVIDIA A100–based. Optimized for large-scale model training and inference. Supports FP64, TF32, and FP16 operations with NVLink and NVSwitch for efficient distributed training and multi-GPU scaling.
-
AI(H100) AMD EPYC 7R13 3.6GHz | NVIDIA H100 80GB * 1 GPU | Only LinuxvCPUs16Memory256Credit / Time 30,800BenchmarkTestSingle-Core Score1680Multi-Core Score6732GPU Score325474NVIDIA H100–based. Designed for large-scale LLMs and multi-node training. Maximizes training speed with FP8/TF32 precision and NVLink, delivering top-tier performance for massive fine-tuning and HPC workloads.vCPUs192Memory2048Credit / Time 272,300BenchmarkTestSingle-Core Score1709Multi-Core Score16009GPU Score334465NVIDIA H100–based. Designed for large-scale LLMs and multi-node training. Maximizes training speed with FP8/TF32 precision and NVLink, delivering top-tier performance for massive fine-tuning and HPC workloads.
-
AI(H200) Intel Xeon Scalable (Sapphire Rapids) 2.7GHz | NVIDIA H200 141GB * 8 GPU | Only LinuxvCPUs192Memory2048Credit / Time 312,200BenchmarkTestSingle-Core Score2048Multi-Core Score20489GPU Score384329NVIDIA H200–based. Next-generation GPU optimized for large-scale generative AI training and inference. Equipped with 141 GB of HBM3 memory and up to 4.8 TB/s bandwidth, supporting FP8, FP16, and BF16 operations for maximum efficiency in large-model workloads.