NVIDIA has unveiled the pricing and expected launch timeline for its next-generation professional GPUs. The flagship RTX PRO 6000 Blackwell GPU is listed by US-based IT reseller Connection at $8,435.44 to $8,565.18. The GPUs, announced at GTC 2025, are set to launch in May. The RTX PRO 6000 comes equipped with 96GB of VRAM and will be available in a Workstation Edition and a Max-Q subvariant.
The Max-Q version is designed to consume less power at 300W TDP while maintaining 88% of the AI performance. Connection has also listed prices for other NVIDIA GPUs in the lineup:
– RTX PRO 5000 (48GB VRAM): $4,439.50 to $4,569.24– RTX PRO 4500 (32GB VRAM): $2,493.42 to $2,623.16– RTX PRO 4000 (Small Form Factor): Starts at $1,481.46– RTX PRO 2000: $722.49
Interestingly, the retailer has included RTX 2000 Blackwell model plans, which NVIDIA has not officially announced. They have also listed a non-SFF version of the RTX 4000 despite no official word from NVIDIA about such a variant.
This suggests there may be upcoming releases that have not been confirmed. The forthcoming Workstation NVIDIA GPUs include:
– RTX PRO 6000 Workstation Edition– RTX PRO 6000 Max-Q Edition– RTX PRO 5000– RTX PRO 4500– RTX PRO 4000– RTX PRO 2000
With this early reveal of pricing and availability, NVIDIA continues to push boundaries in the professional GPU market, providing solutions tailored for high-performance and specialized computing needs across various industries. NVIDIA has announced the NVIDIA RTX PRO Blackwell series, a new generation of workstations and server GPUs designed for AI, technical, creative, engineering, and design professionals.
The GPUs offer accelerated computing, AI inference, ray tracing, and neural rendering technologies. The RTX PRO Blackwell series targets applications such as agentic AI, simulation, extended reality, 3D design, and complex visual effects. The lineup provides cutting-edge computing power, memory capacity, and data throughput for professionals, whether they are using desktops, mobile workstations, or data center GPUs.
The new lineup includes:
– Data Center GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition– Desktop GPUs: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition, NVIDIA RTX PRO 5000 Blackwell, NVIDIA RTX PRO 4500 Blackwell, NVIDIA RTX PRO 4000 Blackwell– Laptop GPUs: NVIDIA RTX PRO 5000 Blackwell, NVIDIA RTX PRO 4000 Blackwell, NVIDIA RTX PRO 3000 Blackwell, NVIDIA RTX PRO 2000 Blackwell, NVIDIA RTX PRO 1000 Blackwell, NVIDIA RTX PRO 500 Blackwell
Bob Pette, vice president of enterprise platforms at NVIDIA, stated, “Bringing NVIDIA Blackwell to workstations and servers will enhance productivity, performance, and speed for AI inference serving, data science, visualization, and content creation.
The RTX PRO Blackwell GPUs feature cutting-edge technology, including:
– NVIDIA Streaming Multiprocessor: Up to 1.5x faster throughput and new neural shaders for AI-augmented graphics innovations– Fourth-Generation RT Cores: Up to 2x performance for photoreal, physically accurate scenes and complex 3D designs– Fifth-Generation Tensor Cores: Up to 4,000 AI trillion operations per second with support for FP4 precision and NVIDIA DLSS 4 Multi Frame Generation– GDDR7 Memory: Boosted capacity—up to 96GB for workstations and servers, up to 24GB on laptops—enabling faster applications and larger datasets– Ninth-Generation NVIDIA NVENC: Accelerated video encoding speed and improved quality, supporting 4:2:2 encoding– Sixth-Generation NVIDIA NVDEC: Up to double H.264 decoding throughput and support for 4:2:2 H.264 and HEVC decode– Fifth-Generation PCIe: Double the bandwidth over the previous generation, improving data transfer speeds from CPU memory– DisplayPort 2.1: Supports high-resolution displays up to 4K at 480Hz and 8K at 165Hz, enabling seamless multi-monitor setups and high dynamic range– Multi-Instance GPU (MIG): Enables partitioning of a single GPU into multiple instances, maximizing performance and flexibility
Industry leaders have provided positive feedback on the performance of the NVIDIA RTX PRO Blackwell series, highlighting its potential to enhance productivity and deliver stunning visuals across various applications.
Pricing and availability for RTX GPUs
The NVIDIA RTX PRO Blackwell series represents a significant advancement for industries relying on powerful AI and graphics performance, offering professionals the tools to push their workflows to new heights.
Supermicro has announced a new lineup of AI systems powered by NVIDIA’s Blackwell Ultra platform. The launch features the HGX B300 NVL16 and GB300 NVL72 systems, designed for advanced artificial intelligence applications such as AI reasoning, agentic AI, and video inference. The HGX B300 NVL16 System is an 8U platform with 2.3TB of HBM3e memory per system and node-to-node speeds of 800 Gb/s.
The GB300 NVL72 System integrates 72 NVIDIA Blackwell Ultra GPUs and 36 Grace CPUs, offering over 20TB of HBM3e memory. Both systems are equipped with advanced liquid-cooling solutions that can reduce power consumption by up to 40%. Supermicro’s liquid-cooling technology operates with 40℃ warm water in an 8-node rack configuration or 35℃ in a double-density 16-node rack configuration, making it more compatible with existing data center infrastructure and more energy-efficient compared to traditional cooling methods.
The new systems enhance AI FLOPs, expand HBM3e capacity, and offer up to 800 Gb/s direct-to-GPU networking performance. The HGX B300 NVL16’s 1.8TB/s 16-GPU NVLink domain creates an unprecedented unified memory pool that eliminates data transfer bottlenecks common in current AI applications. The GB300 NVL72 system integrates 72 GPUs and 36 CPUs within a single rack, making it an exascale-capable system with over 20TB of HBM3e memory, capable of handling model sizes that previously required multiple racks of hardware.
Supermicro’s alignment with NVIDIA’s latest technological advancements ensures the company remains a leading player in the rapidly evolving AI infrastructure market. The company’s comprehensive service offerings, including planning, design, and on-site installation, differentiate Supermicro as a complete solutions provider rather than just a hardware vendor. Supermicro’s latest AI systems represent a significant technical leap that addresses fundamental constraints in AI infrastructure performance, making them a valuable offering for enterprises looking to scale their AI capabilities.
Photo by; Nana Dua on Unsplash
