Elastic Desktop Service (EDS) Enterprise Edition now offers graphics workstation instances powered by the professional-grade NVIDIA RTX 5880 GPU, available as Graphics Workstation Ultimate and Workstation Pro configurations. Built on the NVIDIA Ada Lovelace architecture, the RTX 5880 combines third-generation RT Cores, fourth-generation Tensor Cores, and next-generation CUDA® Cores with 48GB of graphics memory to deliver exceptional rendering, graphics, and compute performance.
Core advantages
The new Graphics Workstation Ultimate and Workstation Pro specifications are equipped with professional-grade NVIDIA RTX 5880 GPU. They support multiple billing methods, including subscription, monthly usage plans, and pay-as-you-go.
Flexible Specifications
A full range of specifications from 2 GB to 192 GB of GPU memory, with flexible scaling.
Meets the demands of all GPU-accelerated workloads, from lightweight content moderation to large-scale industrial design and massive model training.
Powerful Performance
Enterprise-grade GPUs deliver exceptional speed and efficiency for rendering and other demanding workloads.
Outperforms comparable GPU models by up to 10% in performance, with broad software compatibility.
Clear and Smooth Experience
Available in 11 global regions for low-latency access.
Our proprietary ASP protocol provides adaptive encoding and rendering, ensuring a low-latency, distortion-resistant experience.
Secure and Trusted
Full-link, end-to-end data security ensures data never lands on local devices. Comprehensive protection covers terminal access, data transmission, and network security.
Supports robust security features like anti-screen capture, screen recording audits, and domain/application whitelists to effectively prevent data leakage.
Stable and Reliable
Cloud computer Service Level Agreement (SLA): A service availability commitment of ≥99.975% for each Elastic Desktop Service cloud computer.
Achieves a 99.99% connection success rate, with minute-level migration and recovery.
Cost-Effective
Flexible purchasing options with competitive pricing for comparable GPU memory specifications.
Elastic and flexible, supporting on-demand scaling up or down to meet your needs.
Performance comparison
The RTX 5880 outperforms the L20 and A10 GPUs across all key specifications, making it better suited for scenarios such as graphics rendering, industrial software, architectural visualization, and AI inference.
GPU specification | Architecture | Primary use case | GPU memory (VRAM) | Memory bandwidth | CUDA cores | FP32* | TF32 | FP16 | FP8 |
5880 | Ada Lovelace | Professional graphics, design, & AI | 48 GB GDDR6 | 960 GB/s | 14080 | 69.3 | 138.5 | 277 | 554 |
L20 | Ada Lovelace | AI compute / data center acceleration | 48 GB GDDR6 | 864 GB/s | 11776 | 59.5 | 59.5 | 119 | 239 |
A10 | NVIDIA Ampere | Interactive rendering / virtual workstations | 24 GB GDDR6 | 600 GB/s | 9216 | 31.2 | 62.5 | 125 | - |
*The gold standard for graphics rendering, especially in scenarios requiring high-fidelity visuals (such as AAA games, film-grade CG, graphics rendering, and deep learning).
Product specifications
The CPU for both the Graphics Workstation Ultimate and Workstation Pro models is the AMD Genoa 9T24 (with a maximum frequency of up to 3.7GHz), and all models are equipped with the NVIDIA RTX 5880 GPU. The cloud computer's vCPU count, memory, and VRAM can be flexibly configured based on your specific use case.
As hibernation is not currently supported, cloud computers purchased under a monthly hour-based plan will have a default policy to automatically shut down upon disconnection. You can cancel or modify this policy in the console after purchase. For details on managing policies, see Manage policies. For details on purchasing, see Create a cloud computer.
The following pricing covers the cost of the cloud computer's compute resources only and does not include storage costs. The final price is subject to what is displayed on the purchase page.
The Workstation Pro - Network-enhanced cloud computer is currently in invite-only preview and only supports UI customization for hardware terminals. To request access to UI customization or to inquire about customizing the hardware terminal's appearance, please submit a ticket.
Specification | Monthly price (USD/month) | 360-hour plan (USD/month) | 250-hour plan (USD/month) | Pay-as-you-go (USD/hour) |
Graphics Cloud Computer – 8 vCPUs, 16 GB memory, 4 GB GPU memory | 156 | 105 | 86 | 0.390489 |
Graphics Workstation - Ultimate – 16 vCPUs, 32 GB memory, 8 GB GPU memory | 293 | 195 | 162 | 0.732594 |
Graphics Workstation - Ultimate – 16 vCPUs, 32 GB memory, 12 GB GPU memory | 371 | 375 | 250 | 0.92808 |
Graphics Workstation - Ultimate – 32 vCPUs, 64 GB memory, 12 GB GPU memory | 469 | 473 | 315 | 1.17244 |
Graphics Workstation - Ultimate – 32 vCPUs, 64 GB memory, 16 GB GPU memory | 547 | 352 | 291 | 1.367932 |
Graphics Workstation - Ultimate – 32 vCPUs, 64 GB memory, 24 GB GPU memory | 782 | 528 | 430 | 1.954398 |
Graphics Workstation - Ultimate – 64 vCPUs, 128 GB memory, 24 GB GPU memory | 977 | 645 | 537 | 2.443120 |
Workstation Pro – 32 vCPUs, 64 GB memory, 48 GB GPU memory | 1173 | 782 | 645 | 2.931842 |
Workstation Pro – 64 vCPUs, 128 GB memory, 48 GB GPU memory | 1564 | 1055 | 860 | 3.909286 |
Workstation Pro – 64 vCPUs, 256 GB memory, 48 GB GPU memory | 1955 | 1329 | 1075 | 4.886729 |
Workstation Pro – 64 vCPUs, 256 GB memory, 96 GB GPU memory | 2932 | 1955 | 1613 | 7.330338 |
Workstation Pro – 96 vCPUs, 384 GB memory, 192 GB GPU memory | 5669 | 3714 | 3118 | 14.172444 |
Workstation Pro - Network-enhanced – 32 vCPUs, 64 GB memory, 48 GB GPU memory | 1290 | 860 | 709 | 3.225075 |
Workstation Pro Network-enhanced – 64 vCPUs, 256 GB memory, 96 GB GPU memory | 3225 | 2150 | 1773 | 8.063421 |
Workstation Pro -Network-enhanced – 96 vCPUs, 384 GB memory, 192 GB GPU memory | 6236 | 4086 | 3429 | 15.589737 |
Use cases
The RTX Ada 5880, with performance comparable to the RTX Ada 6000, is commercially available. Its third-generation RT Cores accelerate ray tracing, reducing latency by up to 50% for real-time rendering in Omniverse compared to other GPUs in its class. This makes it ideal for the embodied intelligence industry.
Creative design & media: Supports design software such as Adobe Photoshop, Illustrator, and 3ds Max, covering the entire workflow from conceptual design to detailed drafting. It meets the demands of high-fidelity visual scenarios, including AAA games and film-grade CG.
CAD & engineering: Compatible with a wide range of CAD software, including AutoCAD, SolidWorks, and Catia. It supports engineering modeling and complex drafting, enabling end-to-end collaboration from design to implementation.
Advanced manufacturing & automotive: Supports complex component design software like Siemens NX and PTC Creo. It leverages GPU-accelerated parametric modeling and reverse engineering to meet the demands of high-precision mechanical design and virtual testing for autonomous driving systems.
Industrial simulation (CAE): Compatible with industrial simulation platforms such as Altair HyperWorks and ANSYS. Through modules for multi-body dynamics analysis and topology optimization, it supports the structural design of large equipment and heavy simulation validation under extreme working conditions.
AI development & inference: Supports mainstream AI frameworks like ONNX Runtime, PyTorch, and TensorFlow. Supports leading operator libraries such as NVIDIA cuDNN, TensorRT, and cuBLAS. Supports large language model (LLM) inference frameworks like SGLang and vLLM. Combined with the low-latency computing power of a high-performance workstation, it enables real-time AI inference at the edge.
Embodied intelligence & digital twins: Supports embodied intelligence simulation platforms like NVIDIA Omniverse, Isaac Sim, and Isaac Lab. By combining cross-domain collaboration with advanced physics engines, it supports robotics development, digital twins, and virtual testing for autonomous driving, accommodating the entire workflow from algorithm validation to deployment.
Supported regions
Graphics Workstation Ultimate and Workstation Pro are available in the following regions:
China (Beijing)
China (Shanghai)
China (Hangzhou)
China (Shenzhen)
China (Chengdu)
China (Guangzhou)
China (Qingdao)
China (Ulanqab)
China (Hong Kong)
Singapore
Japan (Tokyo)
FAQ
Q: I am working on an embodied intelligence project that requires processing complex algorithms and real-time data. What kind of GPU is suitable for me?
A: EDS provides powerful computing capabilities and flexible resource configurations. The EDS-exclusive RTX Ada 5880 graphics card is ideal for the complex algorithms and real-time data processing demands of embodied intelligence. Furthermore, it offers a graphical Ubuntu OS interface, which is well-suited for development in this field.
Q: Can EDS be used for simulation projects?
A: Yes. EDS has high-performance computing capabilities and supports a wide range of simulation software, enabling you to efficiently complete tasks in fluid dynamics, structural analysis, electromagnetic field simulation, and more.
Q: I regularly use tools like AutoCAD and SolidWorks that require a GPU. What product should I choose?
A: We recommend EDS. It is compatible with all major CAD software, including AutoCAD and SolidWorks, helping designers easily manage every step of the workflow, from conceptual design to detailed drafting.
Q: Our company develops autonomous driving technology and needs high-spec GPUs. What product is right for us?
A: EDS is highly recommended. Its powerful computing capabilities support the research and development of autonomous driving technology, including key stages like sensor fusion, path planning, and machine learning model training.
Q: Our company produces digital human livestreams. What product should we choose?
A: EDS is the perfect choice. It can handle complex video encoding and streaming tasks and allows a single user to run multiple accounts simultaneously, making it ideal for setting up automated livestreaming studios. Its powerful graphics processing also makes it the ideal platform for digital human livestreaming, ensuring the smooth rendering of high-definition avatars.
Q: Should I choose the RTX 4090 or EDS' RTX Ada 5880?
A: The RTX 4090 is better suited for gaming and some light professional graphics work. In contrast, the EDS RTX Ada 5880 is designed for professionals and excels in fields like CAD design, scientific simulation, and deep learning research. The 5880 offers 48 GB of VRAM or more, enabling it to better handle large datasets and complex models.
Q: What is special about the graphics processing and computing performance of EDS GPUs?
The EDS graphics workstation is based on the RTX Ada 5880 and utilizes the latest NVIDIA GPU architecture. It delivers outstanding graphics rendering speed and parallel computing power, making it better suited for high-performance applications like deep learning and 3D modeling.
Q: What cost advantages does EDS GPU have over physical GPU servers?
A: Its pay-as-you-go model lowers initial investment costs and eliminates the hassle of hardware maintenance and upgrades. This makes it especially suitable for businesses and individual users who have limited budgets but high performance requirements.
Q: What types of projects or industries are best suited for the EDS graphics workstation?
A: It is ideal for any work that requires intensive graphics processing, such as film and TV post-production, architectural design, scientific computing, and machine learning.
What are the main use cases for the EDS graphics workstation (RTX Ada 5880)?
A: EDS offers various GPU specifications to fit different business needs. The EDS-exclusive RTX Ada 5880 is particularly well-suited for the following scenarios:
Embodied intelligence: Ideal for the complex algorithms and real-time data processing needs, with a visual Ubuntu OS interface perfect for development.
Engineering & simulation: Its high-performance computing and visual interface support various simulation software for tasks like fluid dynamics, structural analysis, and electromagnetic simulation.
Data annotation: Provides stable and efficient computing resources to accelerate data annotation and support the management of large-scale datasets.
Design & rendering: Supports professional design software like Adobe Photoshop and Illustrator. It's the gold standard for graphics rendering, especially in scenarios requiring high-fidelity visuals like AAA games and film-grade CG.
Autonomous driving: Its powerful computing supports key R&D stages, including sensor fusion, path planning, and machine learning model training.
Digital human & livestreaming: Handles complex video encoding and streaming, supports running multiple instances for automated livestreaming, and smoothly renders high-definition avatars.
Deep learning: The 5880's large 48 GB VRAM capacity allows it to handle massive datasets and complex models, making it an ideal platform for deep learning inference tasks.
Online education & development: Supports various programming languages and IDEs. With one-click image replication, it makes online programming and teaching activities more efficient and flexible.
Q: What are the advantages of the EDS graphics workstation (RTX Ada 5880)?
A: The core advantages lie in its high-performance computing, flexibility, cost optimization, global collaboration, and end-to-end security.
High performance & user-friendly OS: Offers both Windows and Ubuntu with efficient, visual interfaces. Powered by the latest NVIDIA GPU architecture, it provides excellent rendering speed and quality for designers, animators, and post-production artists.
Flexible & scalable: Provides a full range of specifications, from 4 GB to 192 GB of VRAM. Users can choose configurations from a single GPU to multiple GPUs, and even build GPU clusters to meet needs from small projects to large enterprise applications.
Cost-effective: Offers more competitive pricing than comparable physical graphics cards. With flexible payment models (pay-as-you-go, monthly hour-based plans, unlimited monthly plans), it's a great option for startups and individuals with limited budgets.
Global collaboration: With low-latency access points in 11 regions worldwide, users can enjoy a fast, stable connection from anywhere and easily collaborate with international teams, all powered by Alibaba Cloud's global infrastructure.
End-to-end data security: Provides a secure computing environment where all data is stored in the cloud, with no data remaining on local devices. Combined with built-in security features and multiple backup mechanisms, it ensures data integrity and business continuity.
High availability: Allows a single user to run multiple accounts/instances simultaneously.
Business continuity: Local network or power outages do not affect business operations running in the cloud.