Hardware-accelerated compute.
Xilinx and Intel FPGA instances for custom hardware acceleration. Accelerate genomics, video transcoding, financial computing, and network appliances with reconfigurable logic.
Xilinx & Intel
FPGAs
Runtime partial
Reconfiguration
Vitis / Quartus
Development
100 Gbps
Network
Reconfigurable acceleration.
Xilinx and Intel FPGAs for custom hardware acceleration workloads.
Xilinx and Intel FPGAs
Alveo U250/U280 and Intel Agilex FPGAs attached to high-memory instances. Pre-built bitstreams for common workloads.
Runtime reconfiguration
Partial reconfiguration allows loading new accelerator functions without restarting the instance.
Development tools
Vitis and Quartus development environments pre-installed. High-level synthesis from C++ and OpenCL.
AMI marketplace
Pre-built FPGA images for genomics (GATK), video (HEVC/AV1 encode), and financial (options pricing) acceleration.
SmartNIC acceleration
Use FPGAs as programmable SmartNICs for custom network processing, packet filtering, and encryption offload.
Deterministic latency
Hardware execution paths deliver microsecond-level deterministic latency — no OS jitter, no garbage collection pauses.
Getting started
Launch your first instance in three steps. CLI, console, or API — your choice.
ur compute instances create fpga-dev \
--machine-type=f1-standard-16 \
--accelerator=type=xilinx-u280 \
--image=fpga-developer-amiHardware-speed workflows.
Genomics, financial computing, and video encoding — 10-100× faster in hardware.
Accelerated genome analysis
GATK variant calling 30× faster with FPGA acceleration. Process whole-genome samples in under 30 minutes.
View tutorialSuggested configuration
F1-standard-16 · Xilinx U280 · GATK
Estimate your costs
Create detailed configurations to see exactly how much your architecture will cost. Pay for what you use, down to the second.
Configuration 1
Platform & Architecture
Compute Resources
Storage
Cost Optimization
Cost details
Custom bitstream deployment. Shell images for common workloads.
Works seamlessly with
Frequently asked questions
Accelerate in hardware.
Launch FPGA instances with pre-built accelerators or develop your own custom logic.