Intel Gaudi 3 HLB-325 Baseboard vs Supermicro X14 Gaudi 3 AI Training Platform
Accelerator configuration, aggregate memory, bandwidth and power, side by side.
Specifications
| Specification | Intel Gaudi 3 HLB-325 Baseboard | Supermicro X14 Gaudi 3 AI Training Platform |
|---|---|---|
| Vendor | Intel | Supermicro |
| Part number | HLB-325 | — |
| Form factor | baseboard | rack |
| Released | 2024 | 2024 |
| Accelerator | 8x Gaudi 3 | 8x Gaudi 3 |
| Aggregate memory | 1.0 TB | 1.0 TB |
| Aggregate bandwidth | — | — |
| System power | 7.6 kW | — |
| Interconnect | Direct all-to-all between the eight accelerators at 4.2 TB/s bidirectional with no switching IC, plus 1.2 TB/s bidirectional scale-out over 6 OSFP connectors as 24x 200 Gbps RDMA (RoCE v2); host attach is 8x PCIe Gen 5.0 x16 | Eight Gaudi 3 accelerators all-to-all on an OAM 2.0 universal baseboard, with six on-board OSFP 800GbE ports for scale-out over standard Ethernet |
Rows shown in grey are identical between the two.
Intel Gaudi 3 HLB-325 Baseboard
Intel's eight-accelerator universal baseboard for Gaudi 3, and the unit Intel offers OEMs to build servers around. It carries eight HL-325L OAM mezzanine cards and connects them all-to-all at 4.2 TB/s bidirectional without a switching chip, which is the design's distinguishing choice: scale-up needs no switch silicon, and scale-out runs over standard Ethernet at 1.2 TB/s through six OSFP connectors rather than a proprietary fabric. Intel rates the whole assembly at 7.6 kW including the eight OAMs, draws it from a 54 V main input with 12 V standby, and sizes the board at 417 by 585 mm to drop into 19-inch server designs. Sold either as the bare HLB-325 or, as the HLFB-325L, pre-populated with eight cards and tested as one subsystem. Intel publishes no aggregate throughput figure for the baseboard, so none is shown: per-accelerator compute is on the Gaudi 3 chip entry, sourced to Table 5 of Intel's technical paper. Total memory of 1.024 TB is the eight cards' published 128 GB apiece.
Vendor product pageSupermicro X14 Gaudi 3 AI Training Platform
Supermicro's X14 platform for Intel Gaudi 3, and by Supermicro's account the first AI server built on the accelerator. Eight Gaudi 3 cards sit on an OAM 2.0 universal baseboard driven by two Intel Xeon 6 processors, with 24 DIMM slots taking up to 6 TB of host DDR5 in one-DIMM-per-channel, up to eight hot-swap PCIe 5.0 NVMe drives, and four further PCIe 5.0 slots. Scale-out is six on-board OSFP 800GbE ports over ordinary Ethernet rather than a proprietary fabric, which is the Gaudi platform's central pitch alongside an open software stack with no licensing cost. Power is eight 3,000 W Titanium supplies in a 4+4 redundant arrangement, which is supply capacity rather than a draw figure, so no system power is recorded. Air-cooled builds use the HL-325L accelerator and liquid-cooled builds the HL-335, which is a part this catalogue does not yet carry. Supermicro publishes no throughput figure for the system, so none is shown; per-accelerator compute is on the Gaudi 3 chip entry. Supermicro describes the baseboard only as an OAM 2.0 or OAM universal baseboard and never names Intel's HLB-325, though that is what the description matches. Search results give the part number as SYS-822GA-NGR3 in an 8U chassis, but neither Supermicro page carrying that detail could be loaded, so it is reported here rather than recorded as a specification.
Vendor product page