NVIDIA GB300 NVL72

Headlines · 2

1. Oct 23
   vLLM Reports 954K Tokens per Second per GPU for DeepSeek V4.1 Flash

   In a vLLM-published benchmark, DeepSeek V4.1 Flash reached 954,000 tokens per second per GPU on a single NVIDIA GB300 NVL72 rack using P/D disaggregation and Wide-EP, with P90 throughput of 196 tokens per second per user. On GB200, vLLM’s decode throughput was within 6% of DeepInfer’s published results across batch sizes 8–256. A specialized decode path achieved 826 P90 tokens per second for one user on four B300 GPUs. In an AgentX prefill test at concurrency 256, DEP2 delivered 64,400 tokens per second per GPU, compared with 53,300 for DEP4. The report said the configurations differed and did not characterize the difference as an engine-to-engine speedup; replay tests attributed a 13.6% DEP4 throughput loss to data-parallel imbalance.

   vLLM Blog · 🔥 0
2. Oct 14
   Bitdeer adds 67 MW Malaysia AI data center with no customers signed

   Bitdeer announced an October 8 ten-year agreement securing 67 megawatts (MW) of AI data center capacity in Malaysia for A901, its third Malaysian hub, designed for liquid-cooled Nvidia GB300 NVL72 systems. The facility is expected to start delivering capacity in the first quarter of 2027 and reach its full 67 MW in the second quarter, but Bitdeer reported no signed customer commitments for A901 at announcement. A separate 60 MW A103 project announced October 12, which expands the Cyberjaya campus for next-generation Nvidia Vera Rubin systems, lifted Bitdeer's secured global AI cloud capacity to roughly 333.5 MW, about 95% of its target of up to 350 MW by the first quarter of 2028.

   TechRepublic · 🔥 0

Experience and discussion from the community

Share my NVIDIA GB300 NVL72 experienceAsk about NVIDIA GB300 NVL72