MLPerf Inference v6.1 introduces the first agentic workload benchmarked on datacenter hardware and the first model exceeding one trillion parameters, according to Lambda1. The release also recorded 8.85% higher throughput on identical hardware compared with v6.0.
MLPerf Inference v6.1 Adds First Agentic Workload, Trillion-Parameter Model
MLPerf Inference v6.1 introduces the first agentic datacenter workload, the first trillion-parameter model, and 8.85% throughput gains over v6.0.