130 parts · 1,202 companies
Prices as of2026-10-25
Search
4.03

Who makes GPU baseboards & compute trays for AI data centers

GPU baseboards & compute trays (HGX, Bianca, UBB/OAM)

The baseboard is the printed circuit assembly that carries the GPUs (and often paired CPUs) and their direct power and interconnect, packaged into a tray that slides into the rack.

The compute tray sits between the packaged silicon (layer 1, layer 2 memory) and the rack-scale system (4.01): it's the board that physically carries the GPUs, any paired CPUs, VRMs (4.08), connectors (4.07) and the traces that bring scale-up fabric (3.01) to each chip. Nvidia's HGX and Bianca designs and the OAM/UBB open standard define the mechanical and electrical envelope; the ODMs build to that spec while Supermicro builds boards in-house rather than sourcing from the ODM tier. AI demand is the entire reason this part category exists in its current form — a baseboard carrying multiple high-power GPUs, coherent CPU links and dense high-speed interconnect is a different design problem from a general-purpose server motherboard, with far tighter power delivery, signal integrity and thermal requirements. For investors, the baseboard is where board complexity, layer count (4.05) and connector density all stack up, so it's a useful proxy for how much of a rack's bill of materials sits in high-spec board manufacturing versus commodity components; watch which board builders get first-wave qualification on each new GPU generation, since that tends to set the pecking order for the whole rack cycle.

AI delta: A compute tray carrying multiple high-power GPUs with coherent CPU links and dense scale-up interconnect is a fundamentally different design than a general-purpose server motherboard, which is why this category exists as its own part class.

Per rack: 18 compute trays per NVL72 (Nvidia rack specification, via taxonomy v0.1 Appendix A).

Appears at:Enterprise / small coloLarge colo / hyperscale buildingAI campusGW-class campus· cooling:any — present in every GPU server regardless of scale; what changes is tray count per rack and whether the tray is air-cooled or fitted with cold plates.

Companies on this part (8)

CompanyRoleOwnershipTickerPrice1dConfidenceNote
Foxconn (Hon Hai)manufacturerpublic2317.TW245 TWD-1.6%inferredboards built by
Inventecmanufacturerpublic2356.TW60.50 TWD-1.0%inferred
NvidiamanufacturerpublicNVDA211 USD-0.7%disclosedand AMD designs · source
Pegatronmanufacturerpublic4938.TW104 TWD-1.0%inferred
Quanta Computermanufacturerpublic2382.TW322 TWD-1.8%inferred
Supermicroin housepublicSMCI32.94 USD-1.7%inferredin-house
Wistronmanufacturerpublic3231.TW190 TWD-2.6%inferred
Wiwynn · unit of Wistronmanufacturerpublic6669.TW1840 TWD-2.6%inferred

Who competes here

The same table read the other way: every company above competes for GPU baseboards & compute trays sockets. Confidence tags matter — an inferred placement is a lead, not a confirmed rivalry.

Seen in real machines

GPU baseboards & compute trays is named in 17 of the rack teardowns and supercomputer breakdowns on this site. Each line below is a piece of that machine. The companies shown are the ones named on that line, not necessarily the maker of this part.

Nearby parts

From the blog

Latest feed

Questions about GPU baseboards & compute trays

What are GPU baseboards & compute trays?
The baseboard is the printed circuit assembly that carries the GPUs (and often paired CPUs) and their direct power and interconnect, packaged into a tray that slides into the rack.
Which companies make GPU baseboards & compute trays for AI data centers?
This catalog lists 8 companies on GPU baseboards & compute trays, 8 of them with listed equity: Foxconn (Hon Hai), Inventec, Nvidia, Pegatron, Quanta Computer, Supermicro, Wistron, Wiwynn.
How does AI change demand for GPU baseboards & compute trays?
A compute tray carrying multiple high-power GPUs with coherent CPU links and dense scale-up interconnect is a fundamentally different design than a general-purpose server motherboard, which is why this category exists as its own part class.
How many GPU baseboards & compute trays does a data center need?
Per rack: 18 compute trays per NVL72 (Nvidia rack specification, via taxonomy v0.1 Appendix A). Figures come from the taxonomy behind this catalog.
How reliable are these GPU baseboards & compute trays supplier placements?
The 8 placements on this part break down as 1 disclosed, 7 inferred. "disclosed" means the company or a filing says so; "reported" comes from trade press or a reference-design list; "inferred" is a taxonomy lead with no citation yet.

Sources: placements without a source URL come from the taxonomy seed and are tagged accordingly. How this data is built.· Markdown · JSON