Finder new Ranking API

p4de Instance Family

Description Info Performance ranking Available Regions Pricing Charts

The EC2 P4de instance family, launched in preview on May 26, 2022, features 8 NVIDIA A100 GPUs, 96 vCPUs and 1.1 TB of system memory. These instances are designed for machine learning training and high-performance computing. They also offer 8 TB of local NVMe-based SSD storage and 400 Gbps of networking bandwidth, leveraging EC2 UltraClusters for petabit-scale networking.

Source

Instance Types
p4de.24xlarge
vCPUs
Up to 96
Memory
Up to 1152 GiB
Architecture
x86_64
Processor
Intel
GPUs
Up to 8
Ranked by CPU performance (Passmark)
#1
m8a, c8a, r8a
3846 Up to 192 vCPU
#2
m7i, m8i, m7i-flex, m8i-flex, c7i, c7i-flex, c8i, c8i-flex, i7i, i7ie, r7i, r7iz, r8i, r8i-flex, u7i-12tb, u7i-6tb, u7i-8tb, u7in-16tb, u7in-24tb, u7in-32tb
3009 Up to 896 vCPU
#3
m7a, c7a, r7a
2887 Up to 192 vCPU
#4
m6a, m5zn, c6a, r6a
2595 Up to 192 vCPU
#5
m6i, m6id, m6in, m6idn, c6i, c6id, c6in, i4i, r6i, r6id, r6idn, r6in, z1d
2303 Up to 128 vCPU
#6
m5d, m8g, m8gb, m8gd, m8gn, c5, c5a, c5ad, c5d, c5n, c8g, c8gb, c8gd, c8gn, i8g, i8ge, r8g, r8gb, r8gd, r8gn, x8g
1853 Up to 192 vCPU
#7
m5dn, c4, r5dn, r5n, t2
1764 Up to 96 vCPU
#8
m4, m5, m5n, m7g, m7gd, c3, c7g, c7gd, c7gn, i3, i3en, r3, r4, r5, r5b, r5d, r7g, r7gd, t3
1648 Up to 96 vCPU
#9
m5a, m5ad, r5a, r5ad, t3a
1419 Up to 96 vCPU
#10
m6g, m6gd, c6g, c6gd, c6gn, i4g, r6g, r6gd, t4g
1098 Up to 64 vCPU
#11
m1, m2, m3, mac1, mac2, mac-m4, mac2-m2, mac-m4pro, mac2-m2pro, mac2-m1ultra, a1, c1, d2, d3, d3en, dl1, f1, f2, g4ad, g4dn, g5, g5g, g6, g6e, g6f, gr6, gr6f, h1, hpc7g, i2, im4gn, inf1, inf2, is4gen, p3, p3dn, p4d, p4de, p5, p5en, p6-b200, t1, trn1, trn1n, u-3tb1, u-6tb1, vt1, x1, x1e, x2gd, x2idn, x2iedn, x2iezn
Unknown Up to 448 vCPU
🇯🇵 ap-northeast-1
🇸🇬 ap-southeast-1
🇩🇪 eu-central-1
🇮🇱 il-central-1
🇺🇸 us-east-1
🇺🇸 us-west-2

Atom feed for new regions

p4de.24xlarge 96 vCPU, 1152 GiB

On-Demand Price

Spot Price

Markdown (.md) | JSON (.json)
Home
RunsOn