TOMORROW’S TECH
← Chips & Compute

AMD Instinct MI355X

AMD · #4 most active of 18 in Chips & Compute

Compare →

288GB HBM3E accelerators with the ROCm software stack, used by several inference providers.

Current version

MI355X

Entry cost

~$2-3 per GPU-hour on clouds

Changes / 30d

0

Capabilities

  • 288GB HBM3E
  • FP4/FP6 support
  • ROCm 7
  • open rack designs

Pricing

~$2-3 per GPU-hour on clouds

Entry price over time

Not enough pricing history yet — we start charting from the second observation.

Geek mode

Models under the hood

  • CDNA 4

    AMD

    disclosed

  • ROCm

    AMD

    disclosed

Context window

—

Public API

yes

Multi-model routing

no

Stack signals

  • ROCm
  • HIP
  • Infinity Fabric
  • HBM3E

Shared model stack

Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.

green disclosed · amber inferred · grey unknown

Reported scale

Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.

No public usage figure on record for this product.

Also in Chips & Compute

All Chips & Compute
Cerebras WSE-3

Cerebras

1

Wafer-scale inference and training

Version
WSE-3 / Inference
Cost
Free tier, then from $0.06 / M tokens
Model
Llama / Qwen / GPT-OSS
Google TPU Ironwood

Google

1

Inference-first TPU pods

Version
v7 Ironwood
Cost
from ~$1.2 per chip-hour
Model
TPU v7
Groq LPU

Groq

1

Deterministic low-latency inference

Version
GroqCloud
Cost
Free tier, then from $0.05 / M tokens
Model
Llama / Kimi / GPT-OSS
Apple Silicon Neural Engine

Apple

On-device inference across Mac, iPhone and iPad.

Version
M5
Cost
Bundled with hardware
Model
Apple Foundation Models
AWS Trainium2

AWS

Cloud-native training silicon

Version
Trainium2
Cost
from ~$1.3 per accelerator-hour
Model
Trainium2
Broadcom custom accelerators

Broadcom

Co-designed XPUs behind most hyperscaler silicon.

Version
2025
Cost
Enterprise supply agreements
Model
powers Google TPU

Change history

  • version

    MI355X ships with 288GB HBM3E

    MI300XMI355X

    source