TEAS benchmark: measures inference at natural lengths, reporting cost, accuracy, and energy

PontiEdoardo · x · 2026-09-01

Edoardo Ponti introduces TEAS: instead of distorting fixed-length measurements, it uses each dataset's natural lengths, handling chunked prefill, refusals, and steps that mix prefill with decode—capturing real sparse activations and bandwidth utilisation. It reports cost, accuracy, performance (user experience and system output), and energy estimates; accuracy confirms run validity and reveals available trade-offs.

Related event: UK Universities Open-Source TEAS Benchmark for AI Inference Across 9 Accelerators(3 posts)→

Original post →

More from Infra

Infra channel →