ANANSE
No. 1 · Part II
The Ananse Reports: TinyML Reality Check, Part II: The Evidence
The full TinyML matrix: energy, latency, accuracy, variance, and the one result that needed a second look.
Notes from building, and measurements from the bench. Everything published here, in the order it was published.
The full TinyML matrix: energy, latency, accuracy, variance, and the one result that needed a second look.
Three MCU boards, three TinyML workloads, and a measured answer to what AI inference costs in energy on real hardware.
A short tour of the three microcontroller boards and the measurement rig behind the first Ananse Report.
Part of my doctoral research became a practical series: The Ananse Reports, measuring what AI actually costs on real hardware with the method disclosed every time.
Utilizamos cookies analíticos para melhorar sua experiência. Saiba mais