Every shipping NVIDIA accelerator at a glance — memory, bandwidth, interconnect, power, and which workloads each one is built for. Sourced from the official NVIDIA Data Center Platform line card and our own deployment experience.
Built for pretraining, post-training, and large-scale inference. NVLink/NVSwitch-class systems for the biggest models on earth.
| Solution | Networking | Pretraining | Post-Training | Inference / Test-Time | Data Analytics | HPC / Scientific | AI Video Generation |
|---|---|---|---|---|---|---|---|
GB300 NVL72 Blackwell · Rack-scale (72 GPU, liquid) | Quantum-X800 InfiniBand / Spectrum-X | ||||||
GB200 NVL72 Blackwell · Rack-scale (72 GPU, liquid) | Quantum-X800 InfiniBand / Spectrum-X | ||||||
GB200 NVL4 Blackwell · 4-GPU Grace+Blackwell module | Quantum-2 InfiniBand / Spectrum-X | — | — | ||||
GH200 NVL2 / NVL4 Hopper · 2 or 4-GPU Grace+Hopper | Quantum-2 InfiniBand / Spectrum-X | — | — | ||||
HGX B300 (8-GPU) Blackwell · 8-GPU HGX baseboard | Quantum-X800 InfiniBand / Spectrum-X | — | |||||
HGX B200 (8-GPU) Blackwell · 8-GPU HGX baseboard | Quantum-X800 InfiniBand / Spectrum-X | — | |||||
HGX H200 (8-GPU) Hopper · 8-GPU HGX baseboard | Quantum-2 InfiniBand / Spectrum-X | — | — | ||||
H200 NVL (PCIe) Hopper · PCIe dual-slot (NVLink bridge) | Quantum-2 InfiniBand / Spectrum-X | — | — | — |
Versatile PCIe accelerators that span training, inference, rendering, Omniverse, and virtualization in a single SKU.
| Solution | Networking | Pretraining | Post-Training | Inference / Test-Time | Data Analytics | HPC / Scientific | Omniverse / Render | Virtualization (vWS/VDI) | Far Edge / AI Video |
|---|---|---|---|---|---|---|---|---|---|
RTX PRO 6000 Blackwell (Server Edition) Blackwell · PCIe dual-slot | Quantum-2 InfiniBand / Spectrum-X | — | |||||||
L40S Ada Lovelace · PCIe dual-slot | Quantum-2 InfiniBand / Spectrum-3 | — | |||||||
L4 Ada Lovelace · PCIe single-slot, low-profile | Quantum-2 InfiniBand / Spectrum-3 | — |
vGPU is supported across all products shown in this table.
Memory, bandwidth, interconnect, and power for every GPU above.
Flagship rack for frontier reasoning & trillion-parameter training.
Production rack for LLM training and real-time trillion-parameter inference.
Right-sized Grace+Blackwell node for mid-scale training & HPC.
Coherent CPU+GPU memory for HPC, RAG, and graph workloads.
Ultra-class Blackwell baseboard for the largest single-node training jobs.
The new training workhorse — 3× training, 15× inference vs H100 generation.
Hopper refresh: 1.8× LLM inference vs H100 thanks to HBM3e.
PCIe Hopper with HBM3e — best PCIe option for enterprise LLM inference.
Universal Blackwell card: training, inference, rendering, VDI — one SKU.
Best $/perf for generative inference, fine-tuning, and Omniverse rendering.
Low-power universal accelerator for video, edge inference, and VDI.
NVIDIA Quantum-X800 InfiniBand switch + BlueField-3 DPUs and ConnectX-8 SuperNICs. The fabric for Blackwell-class clusters.
NVIDIA Quantum-2 InfiniBand switch + BlueField-3 DPUs and ConnectX-7 NICs. The mainstream 400G AI fabric.
NVIDIA Spectrum-X Ethernet + SuperNICs + BlueField-3 DPUs. Ethernet built for AI workloads.
NVIDIA Spectrum-3 Ethernet + ConnectX-6 NICs + BlueField-2 DPUs. Cost-effective fabric for inference and Omniverse.
1.8 TB/s per-GPU NVLink in Blackwell. Up to 72 GPUs in a single NVL72 domain.
900 GB/s coherent CPU↔GPU link in Grace+Hopper / Grace+Blackwell superchips.
GB300 NVL72 or GB200 NVL72 rack-scale. Liquid-cooled, 5th-Gen NVLink across 72 GPUs, Quantum-X800 fabric.
HGX B200 or HGX H200 8-GPU. Our Scalar L40S is our most-rented inference and fine-tuning platform.
H200 NVL (PCIe) or RTX PRO 6000 Blackwell. 96–141 GB per card with NVLink bridging.
L40S or RTX PRO 6000 Blackwell in dense PCIe servers. Pair with Spectrum-X.
RTX PRO 6000 Blackwell or L40S — the only cards with full RT cores + DLSS in the data center.
L4 in 1U servers — 72 W, single-slot, NVENC × 4 per GPU.
Pick a platform, dial in GPU count, CPU, memory, storage, and fabric. We quote every custom order within one business day.