Data Center AMD

AMD Instinct MI355X

CDNA4 Architecture · 288GB HBM3e · OAM

VRAM
288GB
FP16
-
TDP
500W
Hardware Price
$25k
Cloud from
-
0 providers

Quick Insights

Performance/Dollar
N/A
FP16 performance per $1000
VRAM/Dollar
11.5 GB/$k
VRAM per $1000
vs Data Center Average
N/A
FP16 TFLOPS comparison
Cloud Availability
Not available
Check providers

Specifications

VRAM 288GB HBM3e
Memory Bandwidth 8.0 TB/s
FP16 TFLOPS -
Tensor TFLOPS -
FP32 TFLOPS -
TDP 500W
Form Factor OAM
Architecture CDNA4
NVLink No
Release Date 2025-06

Cloud GPU Pricing

No cloud pricing data available for AMD Instinct MI355X yet.

Browse All Providers →

MI355X vs Alternatives

Compare AMD Instinct MI355X with similar GPUs from other brands.

GPU VRAM FP16 TFLOPS Bandwidth Hardware Price Cloud Price
MI355X Current 288GB - 8.0 TB/s $25k - -
A100 40GB NVIDIA 320GB (+11%) - - - $3.40/hr Compare
B200 NVIDIA 192GB (-33%) - 8.0 TB/s $45k (+80%) $3.75/hr Compare
B200 SXM6 NVIDIA 180GB (-37%) - - - $4.99/hr Compare

Best Use Cases

No specific use case recommendations for AMD Instinct MI355X yet.

Browse All Use Cases →

Compare MI355X

Other AMD GPUs
Alternatives
Add to Comparison Tool

Frequently Asked Questions about MI355X

The AMD Instinct MI355X has a market price of approximately $25k. Cloud rental starts at varies by provider. Prices may vary based on retailer, region, and availability.

Yes, the AMD Instinct MI355X with 288GB VRAM is suitable for many AI/ML workloads. For large language models, you may need multiple GPUs or consider higher-VRAM options like A100 or H100.

Consider buying for long-term, heavy usage (>4 hrs/day). Rent from cloud providers for short-term projects, experimentation, or when you need to scale quickly.

With 288GB VRAM and - FP16 TFLOPS, the AMD Instinct MI355X can run: Large language models (7B-13B), Stable Diffusion XL, video AI, and professional 3D rendering.

The AMD Instinct MI355X offers 288GB VRAM and - FP16 performance at $25k. Compare with similar GPUs using our comparison tool above. Key factors: VRAM for model size, TFLOPS for speed, and price for budget.