EdgeFit Atlas

Measured on-device inference. Every number reproducible; every failure recorded.

mobilenet_v2_1.0_224

hf:google/mobilenet_v2_1.0_224 · vision

Graph fingerprint

not captured

Latency by recipe

CPUaccelerator
024p50 latency, msALL · sm86500.36ALL · sm8650: 0.36 — cv 7.9%, 2 repeats spanning 0.35–0.38 msALL · google-tensor-g45.66ALL · google-tensor-g4: 5.66 — cv 29.2%, 2 repeats spanning 4.71–6.61 ms
Median of 97 timed runs after 3 discarded warmups. Hover a bar for variance and artifact size.

Repeatability

Recipes measured more than once on this unit, in separate sessions. This is the weakest useful form of the check — the real one runs the same recipe on two physical units of the same SKU, and we have one machine. Agreement here cannot catch a defect in the methodology that both runs share; it can only catch drift.

Recipesessionsfastest msslowest msspread
ALL · sm865020.350.385.63%
ALL · google-tensor-g424.716.6140.42%

Every recipe measured

RecipeTargetOutcomep50 mscvsize MiBcosineFLOP fb (auth)time fb (run)partitions
ALLALLsuccess0.357.9%———0.0%—
ALLALLsuccess0.3813.1%———0.0%—
ALLALLsuccess4.7129.2%———0.0%—
ALLALLsuccess6.6114.2%———0.0%—

Reproduce

Commands for every row on this page
uv run edgefit measure-remote --model hf:google/mobilenet_v2_1.0_224 --device "Samsung Galaxy S24 (Family)" --compute-unit all
uv run edgefit measure-remote --model hf:google/mobilenet_v2_1.0_224 --device "Samsung Galaxy S24 (Family)" --compute-unit all
uv run edgefit measure-remote --model hf:google/mobilenet_v2_1.0_224 --device "Google Pixel 9" --compute-unit all
uv run edgefit measure-remote --model hf:google/mobilenet_v2_1.0_224 --device "Google Pixel 9" --compute-unit all