DeepSeek-V3 trained in ~2 minutes (MLPerf v6.0 record)

CoreWeave / NVIDIA — CoreWeave set new MLPerf Training v6.0 records, training DeepSeek-V3 (671B parameters) in 2.02 minutes on 8,192 NVIDIA GB300 NVL72 GPUs — the largest GB300 cluster submitted in the round and the only one scaled beyond 2,048 GPUs on DeepSeek-V3. The run used the same infrastructure customers run in production, a marker of how fast large-model training time is collapsing.