AMD

AMD · CDNA 2 · 2021

AMD MI210.

Rent the AMD MI210 for AI training and inference. 64GB HBM2e, 47.9 TFLOPS FP16, deploy in 60 seconds.

64 GB
HBM2e
1.6 TB/s
Bandwidth
47.9
TFLOPS FP16
4× GPU
Infinity Fabric
Available now · 4 regions

Cloud rental

$0.82/hr

On-demand · per-second billing

$0.70$0.95/hr range

Or $0.57/hr reserved (30% off)

Deploy

One command. AMD MI210 in 60 seconds.

Skip the dashboard if you don't need it. SDK, Python, or cURL — copy the snippet, paste your API key, ship.

import Runcrate from "@runcrate/sdk";

const rc = new Runcrate({ apiKey: "rc_live_••••••••••••••••" });

const instance = await rc.instances.create({
  gpu: "mi210",
  region: "auto",
  image: "runcrate/vllm:latest",
});

console.log(`SSH: ssh root@${instance.host}`);

Benchmarks

How AMD MI210 stacks up.

Side-by-side specs vs the closest alternatives. Bars are normalized to the highest value in each metric.

FP16

AMD MI21047.9 TFLOPS
NVIDIA V10031.4 TFLOPS
AMD MI250X95.7 TFLOPS
NVIDIA RTX 309071 TFLOPS

VRAM

AMD MI21064 GB
NVIDIA V10032 GB
AMD MI250X128 GB
NVIDIA RTX 309024 GB

Bandwidth

AMD MI2101.6 TB/s
NVIDIA V100900 GB/s
AMD MI250X3.3 TB/s
NVIDIA RTX 3090936 GB/s

Full specs

AMD MI210 specifications.

Memory

VRAM64 GB
Memory typeHBM2e
Bandwidth1.6 TB/s

Compute

FP3223.9 TFLOPS
FP1647.9 TFLOPS
INT8192 TOPS
Stream processors6,656

Power & form factor

TDP300 W
Form factorPCIe
ArchitectureCDNA 2
Released2021

Cluster

Max GPUs/node4
InterconnectInfinity Fabric
ManufacturerAMD
Available regions4

FAQ

Frequently asked.

Ready when you are

Deploy AMD MI210
in 60 seconds.

Available across 4 regions · per-second billing · no commitments. Ship today.