MLPerf Training
Time to train to a quality target. Workloads: DeepSeek v3, GPT-OSS 20B, Llama 3.1 8B/405B pretraining; Llama 2 70B fine-tuning; DLRM-dcnv2; FLUX.1
v6.0efficiencyactive
Definition
- Version
- v6.0
- Category
- efficiency
- Maintainer
- MLCommons
- Status
- active
- Home page
- https://mlcommons.org/benchmarks/training/
Sources
- MLCommons MLPerf Training Benchmark mlcommons.org https://mlcommons.org/benchmarks/training/
Scores
The notes record no score on this benchmark.
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MLPerf Training", https://aurion.semai.id/benchmarks/mlperf-training-v6-0/, records of October 2026.