Verification

Same answer, byte for byte

Turning “the deployed model matches the validated one” into a test.

The same model, run on a laptop, on a server and on an emulated microcontroller, returns the same words byte for byte. Not "close". Identical to a single-precision computation. For a product that has to be certified or formally reviewed, that turns "the deployed model behaves like the validated one" from an argument into a test that either passes or fails.

The three environments

StageWhat it is
Python · PyTorchThe reference forward pass. This is the thing your data scientist validated.
C99 · zero-heapThe inference engine. No allocator, no syscalls. This is the thing that ships.
Cortex-M3 · QEMU MPS2-AN385The cross-compiled ELF actually executing on an emulated core.

The numbers

CheckResult
Max |Δ| across all three stages3.7e-7
Multi-token parity, 60K demo model48 / 48 exact
Multi-token parity, 944K model16 / 16 exact
Test suite at HEAD146 / 146 green

Verified by tests/test_parity_with_c.py and tests/test_parity_multitoken.py. Every number reproducible from a cold checkout.

Why the multi-token test is the one that matters

A single-forward-pass parity test is easy to pass and easy to fool. Generation is autoregressive: the model's own output becomes its next input, so a discrepancy in the seventh decimal place at token 1 can become a different word at token 40. Multi-token parity runs the whole generation loop in both environments and compares the emitted sequences. 48 of 48 and 16 of 16 exact means the divergence never got a chance to compound.

The same discipline on the EDGE line

The sensing engine gets the same treatment against a harsher bar, because there the artifact is int8 and quantization is where parity claims usually die.

What a parity claim is worth when it is not executed

A cautionary note that belongs on this page rather than in a changelog. A previous export carried the line "parity: C == torch-static (by construction)". The reasoning was sound and the C engine had never actually been run on the test set. When it finally was, it surfaced a buffer overflow: three arrays hardcoded to 24 channels while the channel count was read at runtime, with no bounds check. On a device that is silent memory corruption, not a wrong answer. It sat there for four weeks behind a claim that sounded like a measurement.

That is why the numbers on this page name the test file that produces them. "By construction" is a hypothesis. A green test on the real artifact is a result.

What is not proved here

Questions

What does bit-exact parity mean?

That the Python reference implementation, the C99 engine and the cross-compiled binary running on an emulated Cortex-M3 produce the same output, byte for byte. The maximum absolute difference across all three stages is 3.7e-7, and multi-token generation is exact: 48 of 48 on the 60K model and 16 of 16 on 944K.

Why does bit-exactness matter for a regulated device?

Because certification asks you to show that what ships behaves like what was validated. If the on-device engine is byte-identical to the reference, that becomes a test with a pass or fail rather than an argument about acceptable drift.

Is parity verified on real hardware or in emulation?

In emulation — QEMU MPS2-AN385 running the actual cross-compiled ELF. That proves the arithmetic path. It does not prove timing or energy, and this site does not claim it does. Separately, the 944K model has been run on a physical ESP32-WROOM-32.