Flagship AI/HPC performance
Delivers ~20× Volta speedups and up to 312 TFLOPS for training and inference workloads.
// gpu compute
NVIDIA A100 Tensor Core GPU Server
Flagship AI Power with Up to 80 GB HBM2e & 3rd-Gen Tensor Cores. Ideal for Deep Learning, Inference, HPC & Data Analytics.
- Online in 1 hour when in stock
- 20+ Tier III global data centers
- Unmetered 1/10/25 Gbps bandwidth
- Crypto payments & 24/7 support included
Custom BuildOut Of Stock
Request Sent
Thanks. An engineer will spec your A100 build and reach out shortly.
Something went wrong. Please try again or email sales@redswitches.com.
// a100 pricing
NVIDIA A100 Tensor Core GPU Server Price
Single and multi-GPU A100 builds on bare metal. No setup fee, full root access, and unmetered 1/10/25 Gbps uplinks.
Reservation Requested
An engineer will reach out shortly to confirm availability and next steps.
Bare Metal // Standard Equipment
All Bare Metal Plans Include
Every server ships fully dedicated: no shared resources, no usage meters, no surprises.
- Setup Cost
- Free
- Provisioning
- Instant & Automated
- Access
- KVM, IPMI, Root
- Protection
- DDoS Shield Included
- Cores
- Up to 128 (Dual Socket)
- Memory
- Up to 2TB RAM
- Storage
- Enterprise NVMe & SSD
- Support
- 24/7/365 Human Engineers
OS & Panels
* Select your OS and panel at checkout
// why redswitches for a100
What You Get With Every NVIDIA A100 Server
Every RedSwitches NVIDIA A100 server is single-tenant bare metal with full root and IPMI access, unmetered 1, 10 or 25 Gbps bandwidth, no setup fee, no bandwidth overage and a flat monthly price. Stocked builds are online in about 1 hour, larger configurations up to 8 GPUs per node are built to order, volume and 6/12-month committed-term discounts apply, payments include crypto, and engineers answer 24/7.
- 1 hr
Online When in Stock
Stocked NVIDIA A100 builds are online in about 1 hour. Configurations not in stock are built to order, and an engineer confirms the lead time before you commit.
- $0
Setup Fee, Flat Monthly
No setup fee and a flat monthly price for the whole server. No per-hour meter and no surprise line items, so GPU spend is forecastable.
- Unmetered
Bandwidth, No Egress Bills
Unmetered 1, 10 or 25 Gbps uplinks are included, and whatever port speed you choose you can use all of it: no overage charges and no egress bills, ever. Moving datasets, checkpoints and model weights in and out costs nothing extra.
- 1 tenant
Bare Metal, Full Control
Single-tenant hardware with root and IPMI access. You choose the OS, drivers, CUDA or ROCm version, and the NVLink or MIG layout, and a private VLAN can link your RedSwitches servers.
- Up to 8x
Multi-GPU, Built to Order
Single and multi-GPU nodes, up to 8 GPUs per server with NVLink where the card supports it. Volume discounts on multi-GPU and multi-server orders, plus 6 and 12-month term savings.
- 24/7
Engineers, Not Bots
Live chat, Telegram and email answered by engineers around the clock. Pay by card, PayPal, bank wire or crypto with no KYC, in 20+ Tier III data centers across the EU, US and Asia.
// compared with cloud
RedSwitches A100 Server vs a Typical Cloud GPU Instance
Same accelerator, different economics: what changes when the GPU sits in a dedicated server you control instead of a metered instance.
| RedSwitches A100 server | Typical cloud GPU instance | |
|---|---|---|
| Billing | Flat monthly price per server | Per-hour or per-second metering |
| Bandwidth | Unmetered 1/10/25 Gbps, use the full port, no overage or egress fees | Egress billed per GB |
| Private networking | Private VLAN between your servers on request | Paid VPC and peering constructs |
| Setup fee | $0 | Varies by instance and region |
| Tenancy | Single-tenant bare metal | Shared, virtualised hosts |
| Access | Root and IPMI, your OS and drivers | Hypervisor-managed images |
| Multi-GPU | Up to 8 per node, NVLink where supported, built to order | Fixed instance shapes |
| Payments | Card, PayPal, bank wire, crypto | Card or invoice |
| Support | 24/7 engineers on chat, Telegram and email | Ticket tiers, paid support plans |
NVIDIA A100 Key Specifications
Ampere GA100 silicon with up to 80 GB HBM2e, third-gen Tensor Cores, and MIG partitioning.
- GPU Architecture
- NVIDIA Ampere GA100 with 312 TFLOPS mixed-precision tensor performance, 3rd-gen Tensor Cores
- 40 GB HBM2
- 1.555 TB/s bandwidth, 250 W (PCIe) / 400 W (SXM)
- 80 GB HBM2e
- 1.935 TB/s, 300 W (PCIe) / 400 W (SXM)
- Multi-Instance GPU (MIG)
- Up to 7 isolated GPU partitions per A100
- NVLink/NVSwitch
- Supports 2× NVLink (PCIe) or up to 16-GPU interconnect at 600 GB/s (SXM)
- FP64
- 9.7 TFLOPS
- TF32
- 156 TFLOPS (312 effective)
- FP16/BF16
- 312 TFLOPS (624 effective)
- INT8
- 624 TOPS (1,248 sparse)
- INT4
- 1,248 TOPS (2,496 sparse)
Variants & Memory
Compute Per Precision
// why a100
Why Choose A100
Flagship AI and HPC performance, elastic MIG slicing, and NVLink scaling on single-tenant bare metal.
Elastic and cost-efficient
MIG allows slicing GPU resources across diverse workloads, maximizing utilization.
Scalable interconnects
NVLink and NVSwitch enable multi-GPU scaling up to 600 GB/s for HPC or training clusters.
Versatile memory options
40 or 80 GB HBM2(e) match workload needs from inference to LLM training.
Broad software ecosystem
Compatible with CUDA, TensorRT, MPI, MLPerf, PyTorch, TensorFlow, and HPC tools.
// use cases
Ideal Use Cases
From deep-learning training to real-time finance, see where the A100 changes what your team can ship.
Deep Learning Training & Finetuning
Handles large LLMs and transformer-based networks with swift TF32/BF16 performance.
Multi-Model Inference Pipelines
MIG enables concurrent serving of multiple AI workloads with guaranteed isolation.
Scientific & HPC Applications
GNNs, fluid dynamics, and quantum simulations see dramatic speedups on A100 clusters.
Data Analytics & Large-Scale ETL
Accelerates RAPIDS-based pipelines and real-time analytics with GPU compute.
Low-latency Finance & Real-Time Workloads
World-class STAC-ML benchmarks show A100 excelling in financial inference and modeling
GPU-Powered Repatriation
Replace cloud GPU VMs with predictable bare-metal performance and no bandwidth fees.
Trusted by Enterprise Teams Worldwide
Read all RedSwitches reviews, or see them on Google, HostAdvice, Cryptwerk and Trustpilot.
// faqs
Deep Dive & FAQs
Common questions about A100 form factors, MIG, multi-GPU scaling, and software support.
PCIe vs SXM?
- SXM: Top-tier performance with 400 W TDP and full NVLink/NVSwitch support.
- PCIe: Easier integration into existing racks. Offers similar MIG and software stack compatibility.
Why choose MIG?
Enables up to 7 hardware-isolated GPU instances, ideal for secure multi-tenant or microservice AI deployment.
Is scaling efficient?
Yes, with NVLink or NVSwitch, clusters scale linearly up to 16 GPUs at 600 GB/s.
Power and cooling?
Our racks support up to 400 W GPUs with advanced cooling, suitable for both PCIe and SXM configurations.
Is there cloud-stack support?
Fully compatible with CUDA 11+, TensorRT, MLPerf, NVIDIA Magnum IO, InfiniBand, and popular frameworks.
// need help
Not Sure Exactly What You Need
No problem. Our talented engineers will consult, architect, migrate, manage, and do whatever it takes to help your business grow and succeed.