GS-SYS-07Engineering datasheet

NVIDIA DGX B300

Statusshipping
Verified2026-07-15
Download datasheet (PDF) ↓

DGX B300 is NVIDIA's factory-built Blackwell Ultra appliance: eight B300 GPUs with 2.3 TB of HBM3e, dual Xeon 6776P hosts, eight ConnectX-8 NICs at 800 Gb/s each, and a single NVIDIA support contract in a 10U air-cooled chassis. It replaces DGX B200 at the top of the DGX line for inference and reasoning workloads, trading B200's stronger FP64 for 50% more GPU memory and about 1.5x the dense FP4 throughput. It is also the first DGX deployable in MGX racks, with front-facing I/O and a 54 VDC busbar variant alongside the classic 200-240 V PSU version. Same buyer logic as every DGX: pay the premium for NVIDIA-validated integration and support, or build the equivalent HGX B300 OEM box and keep BOM control.

8x B300 SXM
GPUs · Blackwell Ultra, NVLink 5
2.3 TB HBM3e
Total GPU memory · 8x 288 GB
144 PFLOPS
FP4 inference · NVIDIA marketing figure
14.5 kW
System max power · air-cooled
10U rackmount
Form factor · MGX-rack deployable
8x ConnectX-8
Networking · up to 800 Gb/s InfiniBand or Ethernet per port
Compute01/05
GPUs8x NVIDIA B300 SXM · Blackwell Ultra architecture
FP4 inference144 PFLOPS · NVIDIA figure; dense per-GPU NVFP4 is 15 PFLOPS
FP8 training72 PFLOPS · NVIDIA figure
GPU-to-GPU fabricNVLink 5 · 1.8 TB/s per GPU bidirectional
CPUs2x Intel Xeon Platinum 6776P · 64 cores each, 2.3 GHz
vs. DGX B2001.5x dense FP4, 2x attention throughput · B200 keeps the FP64 edge
Memory & storage02/05
Total GPU memory2.3 TB HBM3e · 8x 288 GB
System memory2 TB default · upgradable to 4 TB with 128 GB DIMMs
OS storage2x 1.92 TB NVMe M.2 · boot
Data cache storage8x 3.84 TB E1.S NVMe SED · ~30.7 TB raw
Networking03/05
Compute fabric NICs8x ConnectX-8 single-port OSFP · up to 800 Gb/s InfiniBand/Ethernet each
NIC-to-GPU pathPCIe Gen6, direct to GPU · NICs live in the GPU tray, all I/O front-facing
In-band management/storage2x BlueField-3 DPU dual-port QSFP112 · up to 400 Gb/s each
Host management1 GbE RJ-45 on-board + 1 GbE BMC
Power & cooling04/05
System max power14.5 kW · NVIDIA user guide
Power supplies12x 3.2 kW, N+N redundant · 200-240 VAC PSU version
DC variant54 VDC busbar version · for MGX rack deployment
CoolingAir-cooled · front-to-back
Airflow1,500 CFM (PSU) / 1,350 CFM (busbar) · at 70% PWM
Operating temperature10 C to 30 C (50 F to 86 F) · inlet; tighter than DGX B200's 35 C ceiling
Physical05/05
Form factor10U rackmount · first MGX-rack-deployable DGX
Height17.4 in (442 mm)
Width19 in (482.6 mm)
Depth35.74 in (904.2 mm)
Weight370.4 lb (168 kg) PSU version · 271.2 lb (123 kg) busbar version
Field notes
  • DGX B300 is air-cooled at 14.5 kW in a 10U chassis. That is a legacy-colo-killer, not a liquid-cooling problem: no coolant loop needed, but very few air-cooled racks deliver 14.5 kW to one node, let alone several, and 1,500 CFM per node makes hot-aisle containment mandatory.
  • The inlet ceiling dropped to 30 C (vs 35 C on DGX B200). Facilities running warm-aisle setpoints for efficiency need to re-check their supply-air temperature before swapping B300 nodes into B200 slots.
  • 144 PFLOPS FP4 is a marketing figure; dense per-GPU NVFP4 is 15 PFLOPS (120 PFLOPS per node). Size clusters on dense numbers and your own model benchmarks, not the headline.
  • The 2.3 TB of HBM3e is the real purchase driver: a single node now holds a ~600B-parameter model in FP8 with KV-cache headroom. If your working set fits in DGX B200's 1.44 TB, the B300 premium buys you little for today's workload.
  • ConnectX-8 at 800 Gb/s means Quantum-X800 InfiniBand or Spectrum-X800 Ethernet switch fabric to exploit it; pairing with a 400 Gb/s fabric halves the scale-out bandwidth you paid for.
  • FP64 regressed versus B200. Teams with HPC-adjacent workloads (CFD, simulation) mixed into their AI cluster should keep B200 or H200 nodes for those jobs rather than assuming newer is faster everywhere.
Questions we get on this part

How much does a DGX B300 cost?

NVIDIA does not publish a list price. Reported baseline pricing is $300,000 to $350,000 as of Q1 2026, with fully configured reseller quotes running $400,000 to $500,000 depending on support term and region. Quotes go through NVIDIA partners.

Is the DGX B300 air-cooled or liquid-cooled?

Air-cooled. NVIDIA's user guide specifies front-to-back airflow of roughly 1,500 CFM and a 14.5 kW maximum draw in a 10U chassis, with a 10 C to 30 C inlet window. No liquid loop is required, but the per-rack power density usually forces high-density colo space anyway.

DGX B300 vs DGX B200: which one?

DGX B300 carries 2.3 TB of GPU memory (vs 1.44 TB), about 1.5x the dense FP4 throughput and 2x the attention throughput, plus 800 Gb/s ConnectX-8 networking. DGX B200 keeps a significant FP64 advantage and a wider 35 C inlet window. Reasoning/inference at scale favors B300; mixed HPC workloads can still favor B200.

DGX B300 vs HGX B300: what is the difference?

Same 8-GPU Blackwell Ultra platform either way. DGX B300 is NVIDIA's fixed-configuration appliance with NVIDIA-direct support; HGX B300 is the baseboard OEMs (Supermicro, Dell, HPE, Lenovo, Aivres) build their own servers around, including liquid-cooled and denser variants NVIDIA does not sell as DGX.