AMD Instinct MI300X Platform

baseboard
8× Instinct MI300X
4th Gen AMD Infinity Fabric, eight fully connected OAM modules on an OCP UBB 2.0 baseboard, 7x 128 GB/s links per GPU, 896 GB/s ring-of-8 aggregate
2023
FP64
654
TFLOPS
FP32
1,307
TFLOPS
Power
6.0
kW Total
Memory
1536
GB Total

AMD's eight-GPU building block for CDNA 3, integrating eight MI300X OAM modules on an industry-standard OCP universal baseboard (UBB 2.0), each linked to the other seven over 128 GB/s Infinity Fabric links for 896 GB/s of aggregate ring bandwidth, plus one PCIe Gen 5 x16 host link per GPU. AMD's platform datasheet gives 1.5 TB of HBM3 (stored as 8 x 192 GB = 1.536 TB) and 5.3 TB/s per GPU, so 42.4 TB/s across the board, with a 750 W maximum board power per GPU (6.0 kW for the eight GPUs, not counting the host). Compute: 2,432 compute units, 155,648 stream processors at up to 2,100 MHz. AMD sells it through solution partners rather than directly. Precision figures are eight times the per-GPU MI300X row, which AMD's one-decimal platform figures round, except FP32 and FP64, where AMD's platform figures are printed more precisely and are stored as printed.

We will point you at suppliers who have it. Free, and no signup.

FP64
654
TFLOPS
FP32
1,307
TFLOPS
TF32
5,230
TFLOPS
FP16
10,459
TFLOPS

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: CDNA 3
Interconnect: 4th Gen AMD Infinity Fabric, eight fully connected OAM modules on an OCP UBB 2.0 baseboard, 7x 128 GB/s links per GPU, 896 GB/s ring-of-8 aggregate

System Specifications

Form Factor: baseboard
Total Power: 6.0 kW
Total Memory: 1.5 TB
Memory Bandwidth: 42.4 TB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
654 TFLOPS 82 TFLOPS —
1,307 TFLOPS 163 TFLOPS —
5,230 TFLOPS 654 TFLOPS —
10,459 TFLOPS 1,307 TFLOPS —
10,459 TFLOPS 1,307 TFLOPS —
20,919 TFLOPS 2,615 TFLOPS —
20,919 TOPS 2,615 TOPS —

Where a vendor states a basis, the figure shown is the dense one, and any figure it publishes is listed separately. Vendors commonly headline the sparse number instead: for NVIDIA's 2:4 structured sparsity that is exactly twice the dense figure, though other vendors' sparse modes do not all follow that ratio.

Powered by AMD Instinct MI300X

This system utilizes 8 × AMD Instinct MI300X 192GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

750W

Per GPU Memory

192 GB

Process Node

5nm FinFET

Architecture

CDNA 3

Documentation & Resources

Flopper Spec Sheet

AMD Instinct MI300X Platform specifications, generated from our database. Printable.

Vendor product page

AMD Instinct MI300X Platform documentation

View ↗

AMD Instinct MI455X GPU datasheet (PID 5158303, LE-93204-00 07/26)

AMD • 2026-07-21

View Document ↗

Typical Use Cases

AI/ML Training
High-Performance Computing
Data Analytics

The AMD Instinct MI300X Platform runs 8× Instinct MI300X GPUs over 4th Gen AMD Infinity Fabric, eight fully connected OAM modules on an OCP UBB 2.0 baseboard, 7x 128 GB/s links per GPU, 896 GB/s ring-of-8 aggregate, delivering 20,919 TFLOPS FP8 dense.

© 2026 Flopper.io - Compare the GPUs Powering AI