The Definitive Guide toAI Data Centers
Ask the GuideAboutAccount
GuideGlossaryB300

B300

NVIDIA's later Blackwell-generation GPU, the memory-expanded step above B200 and a reference point for how per-GPU HBM capacity and rack power are climbing each generation.

Current numbers

80 → 288 GB → ~1 TBHBM capacity per accelerator package: H100 80 GB; B200 180 GB; B300 288 GB (HBM3E); Rubin Ultra ~1 TB HBM4E on a four-chiplet package (roadmap)as of 2026-08 (roadmap in flux) · register ↗
H100 80GB → Rubin Ultra 1TBper-GPU usable HBM capacity trajectory (H100 80GB, B200 SXM 180GB, B300 SXM 288GB, Rubin 288GB HBM4, Rubin Ultra announced at 1TB)as of 2026-08 (roadmap in flux) · register ↗
~288 GBHBM4 per Rubin GPU at ~22 TB/s; trajectory H100 80 GB → H200 141 GB → B200 180 GB → B300 288 GB → Rubin Ultra 1 TBas of 2026-07 · register ↗

← All terms