Skip to main content
VLSI Korea

Semiconductor graph

AMD Instinct MI300A
Methods ↗

7 entities and 8 documented connections shown.

Connections

7 entities · 8 claims
100%

Drag the map or use pan buttons to move. Tab to a node or connection and press Enter to inspect its evidence. Arrows follow the actual claim direction.

STARTING ENTITYDIRECT CONNECTIONSSECOND STEPBased on architecture · SUPPORTEDBased on architectureBased on architecture · SUPPORTEDBased on architectureUses packaging · SUPPORTEDUses packagingUses memory technology · SUPPORTEDUses memory technologyUses packaging · SUPPORTEDUses packagingUses memory technology · SUPPORTEDUses memory technologyBased on architecture · SUPPORTEDBased on architectureUses memory technology · SUPPORTEDUses memory technologyAMD Instinct MI300X · productPRODUCTAMD Instinct MI300XHBM3 · memory technologyMEMORY TECHNOLOGYHBM3AMD CDNA 3 · architectureARCHITECTUREAMD CDNA 3AMD Zen 4 · architectureARCHITECTUREAMD Zen 4NVIDIA H100 SXM · productPRODUCTNVIDIA H100 SXM3D packaging · packagingPACKAGING3D packagingAMD Instinct MI300A · productPRODUCTSTARTAMD Instinct MI300A
CompanyChip / productTechnologyPlace / facility
Select a connection for its evidence
Relationships & evidence (8)
Every displayed connection is a scoped claim. Arrows follow subject to object.
SubjectRelationshipObjectStatusEvidence
AMD Instinct MI300AAMD Zen 4SUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

This is a heterogeneous APU claim, not a relationship to an external CPU supplier.

Full claim history →
AMD Instinct MI300AAMD CDNA 3SUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

This claim covers the GPU-side architecture only.

Full claim history →
AMD Instinct MI300A3D packagingSUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

No external packaging service provider is asserted.

Full claim history →
NVIDIA H100 SXMHBM3SUPPORTEDSource-reported snapshot: 2022-03-22
Why are these connected?

Variant-specific: NVIDIA also describes PCIe H100 configurations with HBM2e.

NVIDIA Hopper Architecture In-Depth

NVIDIA Technical Blog · Tier A · Primary source · supports

NVIDIA specifies H100 SXM5 with 80 GB HBM3 across five HBM3 stacks.

Location: H100 SXM5 implementation

Published
2022-03-22
Retrieved
2026-09-08

Store factual paraphrase, metadata and link only. No source document redistribution or inferred license; no source bytes retained.

Full claim history →
AMD Instinct MI300X3D packagingSUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

No external packaging service provider is asserted.

Full claim history →
AMD Instinct MI300XHBM3SUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

No memory-chip vendor is asserted.

Full claim history →
AMD Instinct MI300XAMD CDNA 3SUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

Architecture designation does not determine the foundry or package supplier.

Full claim history →
AMD Instinct MI300AHBM3SUPPORTEDSource-reported snapshot: 2023-12-06
Why are these connected?

No memory-chip vendor is asserted.

Full claim history →
Find an evidence path

Find a documented path

From AMD Instinct MI300A · up to two relationship steps.

A path retains its original claims. Shared technology does not establish a supply contract.

7 nodes · 8 claims
Scope

Relationships may describe different dates. A path does not establish simultaneous activity. Bounded source-backed collection. Missing or filtered connections do not establish absence. Supported does not mean independently verified.