EdgeFit Atlas

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

distilbert-base-uncased

hf:distilbert-base-uncased · unknown

Graph fingerprint

not captured

Latency by recipe

CPUaccelerator
0102030p50 latency, msunknown34.48unknown: 34.48 — cv 0.2%, 64 MiB, 10 repeats spanning 11.37–51.66 ms
Median of 10 timed runs after 3 discarded warmups. Hover a bar for variance and artifact size.

Speed against numerics

CPUaccelerator
0.960.970.980.9932.53537.5p50 latency, ms · lower is bettercosine vs fp32 referenceunknown — cv 0.2%, 64 MiB, 10 repeats spanning 11.37–51.66 ms
Marker area scales with artifact size. Cosine is a numerics check against the fp32 PyTorch reference, not task accuracy — a model can hold cosine 0.999 and still fail on the slice that matters.

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
unknown1011.3751.66354.19%

Every recipe measured

RecipeTargetOutcomep50 mscvsize MiBcosineFLOP fb (auth)time fb (run)partitions
unknownunknownsuccess11.370.2%640.97210.0%0.0%
unknownunknownsuccess11.470.2%640.99320.0%0.0%
unknownunknownsuccess32.120.1%2531.00000.0%0.0%
unknownunknownsuccess32.120.1%2531.00000.0%0.0%
unknownunknownsuccess33.380.2%2531.00000.0%
unknownunknownsuccess35.580.2%1271.000099.9%100.0%
unknownunknownsuccess35.650.2%1271.00000.0%0.0%
unknownunknownsuccess36.660.2%2531.000099.8%39.5%31
unknownunknownsuccess39.411.5%2531.000099.8%33.7%31
unknownunknownsuccess51.660.2%2531.000022.0%44
unknownunknownlowering failure

Recorded failures

Kept rather than hidden. A recipe that cannot lower is a fact about the toolchain, and it is what teaches the static filter not to propose it again.

unknown lowering failure
the runtime aborted the process (SIGABRT). This is a delegate-level crash, not a Python exception. Last output: l"(%arg1, %0) <{transpose_lhs = false, transpose_rhs = false}> : (tensor<1x128x1xf32>, tensor<768x768xf32>) -> tensor<1x128x768xf32>
/AppleInternal/Library/BuildRoots/b11baf73-9ee0-11ef-b7b4-7aebe1f78c73/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShadersGraph/mpsgraph/MetalPerformanceShadersGraph/Core/Files/MPSGraphExecutable.mm:975: failed assertion `original module failed verification'

Reproduce

Commands for every row on this page
# recipe df2313959f80086e is no longer in the library
# recipe 8588cf9c7e9e20c2 is no longer in the library
# recipe ab65f76749412d46 is no longer in the library
# recipe 00ae25442f006d4d is no longer in the library
# recipe 827f297b37c07e7d is no longer in the library
# recipe 43f71aef6bb9368a is no longer in the library
# recipe dadb529bf92590e8 is no longer in the library
# recipe a22de0814e01023f is no longer in the library
# recipe caf34ac5af367332 is no longer in the library
# recipe 2be8528fe729cdb1 is no longer in the library
# recipe c7458371172a0612 is no longer in the library