Measured on-device inference. Every number reproducible; every failure recorded.
hf:microsoft/resnet-50 · vision
not captured
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.
| Recipe | sessions | fastest ms | slowest ms | spread |
|---|---|---|---|---|
| ALL · sm8650 | 3 | 0.65 | 1.46 | 126.16% |
| Recipe | Target | Outcome | p50 ms | cv | size MiB | cosine | FLOP fb (auth) | time fb (run) | partitions |
|---|---|---|---|---|---|---|---|---|---|
| ALL | ALL | success | 0.65 | 6.7% | — | — | — | 0.0% | — |
| ALL | ALL | success | 1.43 | 2.7% | — | — | — | 0.0% | — |
| ALL | ALL | success | 1.46 | 4.5% | — | — | — | 0.0% | — |
| ALL | ALL | success | 1.87 | 8.9% | — | — | — | 0.0% | — |
| ALL | ALL | success | 62.36 | 4.7% | — | — | — | 100.0% | — |
| ALL | ALL | success | 78.22 | 4.0% | — | — | — | 100.0% | — |
| ALL | ALL | success | 234.65 | 0.8% | — | — | — | 100.0% | — |
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Samsung Galaxy S24 (Family)" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Samsung Galaxy S24 (Family)" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Samsung Galaxy S24 (Family)" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Snapdragon X Elite CRD" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Google Pixel 10" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Google Pixel 9" --compute-unit all
uv run edgefit measure-remote --model hf:microsoft/resnet-50 --device "Samsung Galaxy A14 5G" --compute-unit all