methodology
Built to answer the practical question first.
WhatsMy.AI is meant to answer one question quickly: what kind of AI should this desktop or laptop realistically handle, and how much confidence should you give that answer?
methodology
WhatsMy.AI is meant to answer one question quickly: what kind of AI should this desktop or laptop realistically handle, and how much confidence should you give that answer?
Browser-visible hardware signals like CPU threads, memory hints, GPU path, and exposed accelerator limits.
Short CPU and GPU throughput checks.
A browser-run reference language model that measures first token, steady speed, context stability, and correctness.
Common NVIDIA and AMD desktop GPUs are cross-checked against the open Epoch AI hardware dataset.
Apple M-series chips are matched against open Wikidata records so the site can identify the chip family more reliably and stay conservative when Safari hides RAM.
The top recommendation is the safest AI to try first on that machine.
The score is only a shortcut. The more useful answers are what AI to try, how fast replies may feel, whether bigger inputs fit, and how confident the site is.
TFLOPs-eq is just an internal power estimate for comparing results on this site, not a manufacturer hardware spec.
Browsers do not expose full VRAM, thermals, power state, or native runtime behavior.
Chrome and Edge currently give the strongest shortlist signal. Firefox is supported more conservatively.
You should still validate important purchase decisions in the exact local AI runtime you plan to use.