Inventec P9000G7 (AC) vs QCT QuantaGrid D75H-10U
Both are built on the NVIDIA HGX B300 8-GPU baseboard, so their accelerator performance is identical rather than merely similar. The choice between them is chassis, cooling, expansion and supplier, not FLOPS.
What actually differs
| Specification | Inventec P9000G7 (AC) | QCT QuantaGrid D75H-10U |
|---|---|---|
| Vendor | Inventec | QCT |
| Part number | P9000G7 (AC) | D75H-10U |
| Form factor | rack | rack |
| Released | — | — |
| Accelerator | 8x B300 | 8x B300 |
| Aggregate memory | — | — |
| Aggregate bandwidth | — | — |
| System power | — | — |
| Interconnect | One NVIDIA HGX B300 air-cooled baseboard, with eight 800Gb OSFP connectors for east-west traffic | One NVIDIA HGX B300 8-GPU baseboard; scale-out is eight ConnectX-8 OSFP ports carrying 800G Ethernet or InfiniBand at one port per GPU for GPU-Direct RDMA |
Rows shown in grey are identical between the two.
Inventec P9000G7 (AC)
The Blackwell Ultra generation of Inventec's 10U air-cooled node, in the same 448 by 441 by 850 mm chassis as its B200 sibling but hosted by Intel Xeon 6700 series with E-cores or 6700 and 6500 series with P-cores. It carries one NVIDIA HGX B300 baseboard and eight 800Gb OSFP connectors for east-west networking, one per accelerator. Storage is eight U.2 NVMe bays with four optional more and two M.2. The accelerator power plane moves to 3,300 W supplies in 5+1 redundancy, tighter than the B200 node's 4+2, with a 2,000 W 12 V host plane. Inventec publishes no throughput, memory or system power figure; the linked HGX B300 entry carries NVIDIA's. The GPU count is not stated and is taken from NVIDIA's designation of HGX B300 as an eight-GPU baseboard.
Vendor product pageQCT QuantaGrid D75H-10U
QCT's 10U Blackwell Ultra server, built on the NVIDIA HGX B300 platform with two Intel Xeon 6 processors. Networking is eight ConnectX-8 OSFP ports at 800G, one per accelerator, for one-to-one GPU-Direct RDMA, and there is room for four full-height full-length PCIe Gen 5 slots. QCT quotes 72 PFLOPS of FP8 and 144 PFLOPS of FP4 without saying whether either is dense or sparse. Both match NVIDIA's sparse figures for this baseboard exactly, and neither matches the dense figures of 36 and 108, so they are recorded as sparse. Worth noting QCT calls the FP8 number a training figure, which would normally imply dense; the number itself settles the basis, not the label. Power is twelve 3,200 W supplies in N+N redundancy, which is capacity rather than draw, so no system power is shown.
Vendor product page