Ultra GPU+ Workstation — 8x RTX PRO 6000 Max-Q

The HEDT crossover desktop, bridging consumer and workstation-class performance. Built on AMD's TRX50 platform with the Threadripper 9960X 24-core processor, 64GB DDR5 ECC registered memory, 4TB Gen5 NVMe storage, and RTX 5090 graphics. Features 10G + 2.5G networking with optional 25G/100G SFP28 expansion. Designed for multi-threaded creative production, simulation, AI training, and workflows that demand more cores and memory bandwidth than consumer Ryzen.

Half-width rackmount 100GbE L3 Lite managed switch with 4x 100GbE QSFP28 ports and 8x 25GbE SFP28 ports (12 ports total). Delivers 1,200 Gbps switching capacity with non-blocking 600 Gbps per direction and a 32K MAC address table. Supports QSFP28-to-4x SFP28 breakout cables for flexible 25GbE fan-out. L3 Lite management via QSS Pro with LACP, VLAN, ACL, LLDP, RSTP, QoS, IGMP Snooping, Flow Control, and Forward Error Correction (FEC). Includes 1GbE RJ45 management port. Half-width 1U form factor allows two units side-by-side in a single rack unit. 100W max power draw on 100–240VAC input. Cost-optimized for enterprise IT, high-speed storage networks, video editing, virtualization, and AI infrastructure.

ZFS-based 1U rackmount all-flash NAS with 10 hot-swap 2.5-inch U.2 NVMe PCIe Gen 4 / SATA 6Gbps bays. Powered by a 16-core/32-thread AMD EPYC 7302P processor (up to 3.3 GHz) with 64GB DDR4 ECC RDIMM expandable to 1TB across 12 DIMM slots. Networking includes 2x 25GbE SFP28 SmartNIC ports and 2x 2.5GbE RJ45, with 2x PCIe Gen 4 x16 expansion slots for additional connectivity. Features 3x USB 3.2 Gen 1 ports, dual redundant 550W PSUs, and runs QuTS hero (ZFS-based OS) with inline deduplication, compression, and near-unlimited snapshots. Purpose-built for low-latency file servers, virtualized workloads, data centers, and 4K/8K media streaming.

AI supercomputer powered by the NVIDIA Grace 72-core Neoverse V2 CPU and NVIDIA DGX B300 with 288GB HBM3e. Includes discrete RTX PRO 1000 Blackwell GPU (8GB GDDR7), 496GB LPDDR5X system memory, and 16TB Gen4 SSD storage. Connectivity includes 2x QSFP28 400GbE ports, 4x USB 3.2 Gen 2, 2x USB-C 3.2 Gen 2, 1GbE and 10GbE Ethernet, mini DisplayPort (BMC), and audio I/O. Runs Ubuntu 24.04 LTS with NVIDIA AI Developer Tools. Purpose-built for large-scale model training, fine-tuning, and AI development.

27-inch 5K Retina XDR display with nano-texture glass and tilt- and height-adjustable stand. 2000 nits peak HDR brightness, mini-LED with 2304 dimming zones, P3 + Adobe RGB wide color gamut, 120Hz adaptive sync. Thunderbolt 5 upstream and downstream, 12MP Center Stage camera, studio-quality three-mic array, six-speaker Spatial Audio.

Ruggedized portable SSD for fast field transfers, on-set data shuttling, and backup. IP65 rated with 3-meter drop protection and hardware encryption for secure, durable mobile storage.

High-speed Thunderbolt 3 direct-attached storage for editors and artists who need fast, local access to large project files. Daisy-chainable with RAID support for flexible capacity and redundancy.

Production-grade rack NAS for shared project storage, media asset management, and automated backups. 8-bay capacity with optional 10GbE and Btrfs snapshots for team-scale file serving.

Premium wireless mouse and keyboard combo for professional workstation setups. Multi-device pairing, quiet operation, and ergonomic design for all-day productivity across Mac and PC.

Intelligent rack PDU with per-outlet switching, metering, and remote management for high-density compute deployments. Daisy-chain capable with SNMP monitoring for lights-out rack management.

Enterprise-grade rack UPS with online double-conversion topology for zero-transfer-time power protection. Keeps servers and workstations running clean through outages, sags, and surges.

A color-accurate portable display with integrated Wacom EMR pen for on-set review, field production, and mobile retouching. Tripod-mountable with 100% Adobe RGB coverage for location-critical color work.

A workhorse enterprise display with IPS Black technology and built-in KVM for multi-system workflows. 140W USB-C power delivery and 120Hz refresh make it the go-to monitor for professional compute setups.

The definitive reference monitor for color-critical production — factory-calibrated OLED with a built-in colorimeter and true blacks. Made for colorists, VFX supervisors, and anyone grading to DCI-P3.

Passive copper direct-attach cable for the lowest-latency, lowest-power 100GbE rack interconnects. Zero power draw and sub-5ns latency make it the ideal short-reach link between switches and adapters.

The industry-standard 100GbE network adapter for GPU-accelerated compute, with hardware RDMA and GPUDirect support. Essential for AI training clusters and high-performance storage fabrics.

A versatile 10GbE managed switch for creative production environments with mixed copper and fiber connectivity. Web-managed Layer 2 switching with VLAN, QoS, and LACP for studio and office deployments.

A hybrid switch bridging 25GbE workstation connectivity with 100GbE uplinks in a single 1U chassis. Ideal for mixed-speed environments connecting workstations to storage and compute clusters.

The backbone switch for high-performance compute clusters, delivering 3.2 Tb/s of non-blocking throughput with ultra-low latency. Supports Cumulus Linux and SONiC for flexible open networking deployments.

Apple's most powerful desktop, with 512GB unified memory for loading massive AI models and handling multi-stream 8K timelines. Purpose-built for AI/ML development, 3D rendering, and post-production at the highest tier of Apple Silicon.

The smallest desktop in the fleet — a 5-inch form factor with M4 Pro performance. Perfect for space-constrained desks, edit bays, and lightweight AI inference where compact footprint matters most.

The lightest laptop in the fleet, built for mobile professionals who need all-day battery life without sacrificing Apple Silicon performance. Ideal for on-site client work, corporate events, and portable creative workflows.

A compact ITX desktop built for creators and developers who need RTX 5090 power in the smallest footprint. The Ryzen 9 9900X3D delivers exceptional single-thread performance with 3D V-Cache, paired with 64GB DDR5 and 2TB NVMe storage. Ideal for 3D rendering, video editing, and AI-assisted creative workflows at a standard 120V outlet.

A personal AI supercomputer in a 1-liter form factor — 1 PFLOP of FP4 AI performance on your desk. Purpose-built for local LLM inference, AI model prototyping, and edge AI development without cloud dependency.

The highest raw GPU horsepower in a laptop — an RTX 5090 Mobile with 24GB GDDR7 in an 18-inch chassis. Built for real-time 3D, game development, and GPU-intensive creative workflows that demand peak mobile performance.

16-inch MacBook Pro in Space Black with M5 Max chip — 18-core CPU, 40-core GPU, 16-core Neural Engine. 128GB unified memory, 4TB SSD, nano-texture display. Three Thunderbolt 5 ports, HDMI, SDXC, MagSafe 3. Supports up to four external displays. Apple Intelligence enabled.

Dell's flagship mobile workstation with a modular GPU design and 100+ ISV certifications. Built for creative professionals and engineers who need certified pro graphics in a portable, MIL-STD-810H rated chassis.

ISV-certified mobile workstation with MIL-STD-810H durability and RTX PRO 5000 graphics. Purpose-built for engineers and designers running CAD, simulation, and 3D modeling on the go.

The largest and most expandable mobile workstation in the fleet, with up to 256GB RAM and four M.2 slots. Built for power users in VFX, simulation, and AI who need desktop-class performance on set or on site.

Quad-GPU server for mid-scale AI training, batch rendering, and multi-tenant GPU workloads. Single-socket EPYC 96-core platform with 384GB total VRAM handles parallel GPU jobs at production scale.

Dual-socket Supermicro server with quad RTX PRO 6000 GPUs and 1TB of system memory. Designed for enterprise AI/ML pipelines, large-scale rendering, and high-availability production compute.

Maximum CPU core density paired with a single RTX PRO 6000 Max-Q — built for CPU-bound workflows like massive simulations, fluid dynamics, and compile-heavy software development alongside GPU-accelerated rendering. 96 cores of Threadripper PRO with 35TB of NVMe storage for the largest datasets.

The top-tier consumer-class desktop with maximum storage and expansion capacity. Built in a full tower with the Ryzen 9 9950X3D 16-core processor, 128GB DDR5, 8TB of NVMe storage (4TB Gen5 + 4TB Gen4), 10G networking, and optional 25G SFP28 connectivity. Designed for real-time 3D visualization, game development, large dataset processing, and latency-sensitive creative work.

A mid-tower desktop with expanded memory and 10G networking for professional creative and development teams. Combines RTX 5090 graphics with the Ryzen 9 9950X3D 16-core processor, 128GB DDR5 RAM, and 4TB Gen5 NVMe storage for multi-app, multi-project workflows.

Dual-GPU rack workstation for render farms, distributed simulation, and multi-GPU VFX compositing. 192GB total ECC VRAM across two RTX PRO 6000 Max-Q GPUs in a 5U rackmount chassis with 2800W platinum power delivery. Built for data center deployment.

Single-GPU professional workstation built on the AMD WRX90 platform with 8-channel DDR5 memory and full PCIe 5.0 lane support. The RTX PRO 6000 Blackwell delivers 96GB of ECC VRAM for large-scene VFX, architectural visualization, and scientific computing. ISV-certified for all major professional applications.

Triple-GPU workstation with 288GB total ECC VRAM for the most demanding multi-GPU rendering, AI training, and large-scale simulation workloads. Max-Q power efficiency enables three GPUs in a full-tower chassis with 512GB system memory for massive scene files.

Entry-level GPU server for inference pipelines, render services, and AI-assisted production workloads. Single-socket EPYC with dual RTX PRO 6000 GPUs balances compute density with cost efficiency.

The flagship GPU server — eight RTX PRO 6000 cards with 768GB total VRAM and 256 CPU cores. Built for enterprise-scale AI model training, massive parallel rendering, and the most demanding HPC workloads.