Skip to content

Instantly share code, notes, and snippets.

@Steboss

Steboss/edgeai_true_false.csv

Created July 1, 2026 10:30
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save Steboss/7bc22e0193770289f6c7d9c11767fa5d to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save Steboss/7bc22e0193770289f6c7d9c11767fa5d to your computer and use it in GitHub Desktop.
Download ZIP
what's true and false in edge AI
Raw
edgeai_true_false.csv
Claim Why isn't true Correct claim
It fits because the file fits The file is the weight payload at one precision. Peak memory also includes activations, workspace, and a KV cache that grows linearly with the context Fit is undetermined wihtout context length and runtime overhead
Parameter count equals memory usage The weight payload should be used, it depends on the precision, wegiths are resident, activations and KV come and go Peak is runtime-specific
FLOPS decide latency FLOPs ignore data movement. A decode step and a prefill step over the same weights do identical airthmetic but differ in intensity The number of tokens M is what decide everything about memory and compute-bound cases
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment