NVIDIA A100 8x80GB Baseboard (A100-8x80GB-Baseboard)
P/N: A100-8x80GB-Baseboard
84 805 € (inc. VAT (Spain))
New
NVIDIA RTX Pro 6000 Blackwell Max-Q Workstation Edition 96GB (900-5G153-2200-000)
P/N: 900-5G153-2200-000
9 629 € (inc. VAT (Spain))
EU Delivery time is made within 3-7 days
This is an estimated timeframe. Delivery times may vary depending on logistics and stock availability.
Warranty 2 year
The product is covered by the manufacturer’s standard warranty.
In stock
Guaranteed Safe Checkout:
RTX Pro 6000 Blackwell Max-Q Workstation Edition 96GB. with fast EU delivery and worldwide shipping. Includes official warranty.
Expert support On-line
Our specialist will help you choose the right server components and ensure full compatibility with your system.
Technical Specifications Product
| Weight | 3,42 kg |
|---|---|
| Dimensions | 30,5 × 13,7 × 17 cm |
| Country of manufacture | Taiwan |
| Manufacturer's warranty (years) | 1 |
| Model | RTX Pro 6000 Blackwell |
| Cache L2 (MB) | 128 |
| Cache L1 (KB) | 24576 |
| Process technology (nm) | 4 |
| Memory type | GDDR7 ECC |
| Graphics Processing Unit (Chip) | |
| Number of CUDA cores | 24064 |
| Number of Tensor cores | 752 |
| Number of RT cores | 188 |
| GPU Frequency (MHz) | 1590 |
| GPU Boost Frequency (MHz) | 2288 |
| Video memory size (GB) | 96 |
| Memory frequency (MHz) | 28000 |
| Memory bus width (bits) | 512 |
| Memory Bandwidth (GB/s) | 1792 |
| Connection interface (PCIe) | PCIe 5.0 x16 |
| FP16 performance (TFLOPS) | 110.1 |
| FP32 performance (TFLOPS) | 55.1 |
| FP64 performance (TFLOPS) | 1.968 |
| NVENC / NVDEC number of blocks | 4×NVDEC, 4×NVENC |
| Cooling type | Active (turbine) |
| Number of occupied slots (pcs) | 2 |
| Temperature range (°C) | 0–50 |
| Multi-GPU support | Yes, via PCIe/Sync, but without NVLink |
| Virtualization/MIG support | vGPU support, not MIG |
| SKU | 900-5G153-2500-000 |
| Architecture | Blackwell |
Product description
Direct import from the USA without intermediaries. Official warranty 1-2 years RTX Pro 6000 Blackwell 96GB
The RTX Pro 6000 Blackwell 96GB series from NVIDIA includes three models: Workstation Edition, Max-Q Workstation Edition, and Server Edition. All of them are built on the latest NVIDIA Blackwell architecture and equipped with 96 GB of GDDR7 video memory with ECC. However, each modification is optimised for different usage scenarios, and it is in these details that the key differences lie.
Each graphics card has 24,064 CUDA cores and supports all current graphics and computing APIs, from DirectX 12 and Vulkan to CUDA 12.8. This makes the series suitable for graphics visualisation as well as deep learning, simulation, big data analysis, and generative AI. However, the way in which cooling, power consumption, performance and dimensions are implemented determines its specialisation.
Specifications
- GPU memory: 96GB GDDR7
- GPU architecture: NVIDIA Blackwell
- NVIDIA CUDA cores: 24064
- Tensor cores: 5th generation
- Ray tracing cores: 4th generation
- AI performance: 4000 AI TOPS
- Single-precision performance: 125 TFLOPS
- RT core performance: 380 TFLOPS
- Memory interface: 512-bit
- Memory bandwidth: 1792 GB/s
- System interface: PCIe 5.0 x16
- Display connectors: 4x DisplayPort 2.1b
- Maximum number of simultaneously connected displays: 4x 4096 x 2160 @ 120 Hz, 4x 5120 x 2880 @ 60 Hz, 2x 7680 x 4320 @ 60 Hz
- Video encoders: 4x NVENC (9th generation), 4x NVDEC (6th generation)
- MIG support: up to 4x 24GB, up to 2x 48GB, up to 1x 96GB
- Power consumption: 600W
- Power connector: 1x PCIe CEM5 16-pin
- Cooling system: Throughflow
- Form factor: 5.4″ H x 12″ L, dual-slot, extended height
- Graphics APIs: DirectX 12, Shader Model 6.6, OpenGL 4.6, Vulkan 1.3
- Computing APIs: CUDA 12.8, OpenCL 3.0, DirectCompute
Max-Q
Workstation Edition: maximum power for your workstation
This version is the choice for those who work with resource-intensive 3D scenes, complex CAD design, 8K video or real-time visualisation. The card operates at up to 2617 MHz, delivers 126 TFLOPS (FP32) and comes with active side cooling. Power consumption is up to 600 W, which requires a powerful PSU but delivers maximum performance.
Key advantage: maximum performance without compromise in the workstation class.
How to choose the right version of the RTX Pro 6000 Blackwell 96GB?
- Need maximum performance on a local machine? — Choose the Workstation Edition.
- Looking for a balance between power and thermal package? — Max-Q Workstation is the perfect choice.
- Working on large-scale AI projects or building server infrastructure? — Server Edition is your choice.
All three cards are equipped with the latest NVIDIA technologies, support DisplayPort 2.1, include GDDR7 ECC memory and offer outstanding computational density.
Why should you buy the RTX Pro 6000 Blackwell 96GB from us?
- Direct delivery from the US — no middlemen, no compromises.
- 3-year warranty — reliability from day one.
- Any form of payment: card, bank transfer (with or without VAT), or cryptocurrency (USDT).
- We will help you choose the right version for your task — from rendering to AI.
Order the RTX Pro 6000 Blackwell 96GB right now and get a professional tool that will take your project to the next level.
Product reviews
0
50
40
30
20
10
Only logged in customers who have purchased this product may leave a review.
Product Berchmark
Review and Analysis of the NVIDIA RTX Pro 6000 Blackwell Max-Q Workstation Edition 96GB
The NVIDIA RTX Pro 6000 Blackwell Max-Q Workstation Edition 96GB is a cutting-edge professional graphics solution based on the latest Blackwell architecture. The graphics card is designed for high-end workstations and is focused on handling resource-intensive tasks in the fields of generative artificial intelligence, deep learning, complex 3D modeling, and scientific computing. The Max-Q version stands out with optimized power consumption while maintaining extreme memory density and computational capabilities.
The main feature of this model is the colossal 96 GB GDDR7 video memory with ECC support, which was previously available only in server solutions like the H100. Thanks to the use of the PCIe 5.0 interface and record-breaking memory bandwidth, the card eliminates bottlenecks when working with Large Language Models (LLMs) and massive visualization scenes, while ensuring high energy efficiency within a 300W thermal package.
Comparative Table of Technical Specifications
The table presents a comparison of the RTX Pro 6000 Blackwell Max-Q with its nine closest competitors in the series and related professional solutions, including the previous flagship generation (Ada Lovelace, Ampere) and server accelerators (H100, L40S).
| Model | VRAM (GB) | Memory Type | Bandwidth (GB/s) | CUDA Cores | Tensor Cores | FP16 (TFLOPS) | FP32 (TFLOPS) | FP64 (TFLOPS) | Tensor Perf (TFLOPS) | Interface | TDP (W) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| NVIDIA RTX Pro 6000 Blackwell Max-Q 96GB | 96 | GDDR7 ECC | 1792 | 24064 | 752 | 110.1 | 55.1 | 1.97 | 220.2 | PCIe 5.0 x16 | 300 |
| NVIDIA RTX Pro 6000 Blackwell (Workstation) | 96 | GDDR7 ECC | 1792 | 24064 | 752 | 110.1 | 55.1 | 1.97 | 220.2 | PCIe 5.0 x16 | 600 |
| NVIDIA RTX Pro 6000 Blackwell (Server) | 96 | GDDR7 ECC | 1597 | 24064 | 752 | 126.0 | 62.5 | 1.72 | 250.0 | PCIe 5.0 x16 | 600 |
| NVIDIA RTX 6000 Ada 48GB | 48 | GDDR6 ECC | 960 | 18176 | 568 | – | 91.1 | – | 1457 | PCIe 4.0 x16 | 300 |
| NVIDIA RTX 5880 Ada 48GB | 48 | GDDR6 ECC | 960 | 14080 | 440 | – | 69.3 | – | 1108 | PCIe 4.0 x16 | 285 |
| NVIDIA RTX A6000 48GB | 48 | GDDR6 | 768 | 10752 | 336 | – | 38.7 | 1.21 | 309.7 | PCIe 4.0 x16 | 300 |
| NVIDIA L40S 48GB | 48 | GDDR6 ECC | 864 | 18176 | 568 | 733.0 | 366.0 | – | – | PCIe 4.0 x16 | 350 |
| NVIDIA H100 96GB | 94 | HBM3 | 3900 | 16896 | 432 | 1979 | 989 | 49.0 | 3958 | PCIe 5.0 x16 | 400 |
| NVIDIA A100 80GB | 80 | HBM2e | 1935 | 6912 | 432 | 312.0 | 156.0 | 9.70 | 624 | PCIe 4.0 x16 | 300 |
| NVIDIA RTX Pro 5000 Blackwell 48GB | 48 | GDDR7 | 1344 | 14080 | 440 | 73.7 | 73.7 | 1.15 | – | PCIe 5.0 x16 | 300 |
*Bandwidth — Memory Bandwidth. Values “-” indicate lack of data.
Analysis of RTX Pro 6000 Blackwell Max-Q Technical Specifications
The analysis shows that the NVIDIA RTX Pro 6000 Blackwell Max-Q occupies a unique position in the market, offering an unprecedented balance between performance and energy efficiency:
- Memory Subsystem Superiority: The 96 GB memory capacity places this card in the same league as the flagship H100, significantly outperforming the previous workstation segment leader RTX 6000 Ada (48 GB) and the server-grade L40S (48 GB). The use of GDDR7 memory with a bandwidth of 1792 GB/s provides nearly a two-fold increase in data transfer speed compared to the Ada Lovelace generation (960 GB/s), which is critical for data-intensive tasks.
- Computational Power: With 24,064 CUDA cores, the new model substantially outperforms the RTX 6000 Ada (18,176 cores) and A100 (6,912 cores). This ensures a significant performance boost in rendering tasks and FP32 calculations. Although the H100 retains leadership in pure FP16 computations, the RTX Pro 6000 offers a more versatile architecture for mixed workloads.
- Max-Q Energy Efficiency: The key advantage of the Max-Q version is its 300W TDP. With identical memory capacity and core count as the “senior” 600W version, this model offers the same architectural power at half the power consumption, making it ideal for installation in standard workstations without the need for complex cooling infrastructure typical of servers with H100 (400-700W).
- Future-Proof Technologies: Support for the PCIe 5.0 interface doubles the data exchange speed with the CPU compared to RTX Ada series cards and the A100 (PCIe 4.0), making this model ready for next-generation platforms.
Conclusion
The NVIDIA RTX Pro 6000 Blackwell Max-Q Workstation Edition 96GB is an uncompromising choice for professionals requiring maximum memory capacity and performance in a workstation form factor. It effectively bridges the gap between traditional professional cards (RTX A6000/6000 Ada) and specialized data center accelerators (H100), providing 96 GB of memory and the power of the Blackwell architecture in an energy-efficient design. This is the best solution for tasks where VRAM capacity and bandwidth are limiting factors, but the infrastructure does not allow for the use of server solutions with power consumption above 300-400W.
Product FAQ
The main difference lies in energy efficiency. The Max-Q version is optimized to operate with a thermal design power (TDP) of 300W, whereas the standard version consumes up to 600W. At the same time, the Max-Q retains the same colossal 96GB video memory capacity and CUDA core count, offering an ideal balance of performance for workstations where low heat generation and acoustic comfort are important.
The RTX Pro 6000 Blackwell Max-Q is designed for the most resource-intensive professional workloads. It is ideally suited for training and fine-tuning Large Language Models (LLMs), generative AI, complex 3D rendering at 8K+ resolution, scientific simulations, and data analysis where a huge video memory buffer is required.
The transition to GDDR7 standard memory has secured a bandwidth of 1792 GB/s, which is almost twice as fast as the previous generation. The 96GB capacity allows for loading massive neural networks or 3D scenes entirely, avoiding slow data exchange with system memory, which is critical for calculation speed.
Yes, the card is equipped with a PCIe 5.0 x16 interface. This doubles the data transfer speed between the processor and the video card compared to PCIe 4.0, which is especially noticeable when working with large datasets and in multi-GPU configurations.
Thanks to the Max-Q design and 300W power consumption, this card does not require extreme power supplies (unlike the 600W version) or server racks with reinforced airflow. It is compatible with most professional Workstation-class cases, simplifying the upgrade of existing equipment.
Due to the new Blackwell architecture, an increase in CUDA cores to 24,064 (vs 18,176 in Ada), and the use of faster GDDR7 memory, the card demonstrates a significant performance boost, particularly in FP32 computing tasks and ray tracing.
Yes, the video card supports ECC (Error Correction Code) for video memory. This is a mandatory requirement for professional computing, engineering calculations, and financial simulations, where any bit error can lead to critical failures or incorrect results.
Yes, the compact dual-slot design and reduced power consumption (300W) make the Max-Q version an excellent candidate for installing two or more cards in a single system to multiply computing power, provided the motherboard and power supply allow it.
When purchasing from our store, you receive an official 3-year manufacturer’s warranty. This provides reliable protection for your investment and guarantees stable equipment operation in professional environments.
Payment & Shipping methods
Fast and reliable delivery across the European Union
Estimated transit time: 3–7 days from order confirmation. Worldwide shipping is available for customers outside the EU.
All orders are processed within 24 hours after confirmation. Tracking information is provided as soon as the parcel leaves our logistics center.
Multiple Secure Payment Methods
We accept: Visa, MasterCard, Apple Pay, Google Pay, PayPal, SEPA, iDEAL, Amazon Pay, Bancontact, International Bank Transfers, Klarna via Stripe gateway.
All transactions are encrypted and processed via certified Stripe payment gateway for your security.
Additional Notes
- Delivery times may vary depending on customs clearance and carrier schedules.
- Large or custom-built items may require additional handling time.
- Shipments are insured until delivered to the customer.
- We do not deliver to P.O. boxes or military addresses.
Customers Also Loved
(incl. VAT)
NVIDIA RTX A5000 24GB (900-5G132-1700-000)
P/N: 900-5G132-1700-000
3 828 € (inc. VAT (Spain))
New
NVIDIA RTX 6000 Ada 48GB (900-5G133-2250-000)
P/N: 900-5G133-2250-000
8 682 € (inc. VAT (Spain))
New
NVIDIA A100 8x40GB Baseboard (A100-8x40GB-Baseboard)
P/N: A100-8x40GB-Baseboard
45 231 € (inc. VAT (Spain))
New
New
New
New
New
Reviews
There are no reviews yet.