quantized duplicated precisions.json, which is authoritative for the
quantized/non-quantized partition. default expressed a per-repository
recommendation nothing consumed: the library resolves its own per-platform
default. Both are gone from the schema.
Encode runs on Vulkan, decode stays on XNNPACK. Exported with the
conv1d-as-conv2d lowering, the slice-based STFT magnitude, hoisted
cross-attention K/V and a head-major KV layout. The int8 build quantizes
every decode-side linear; encode is identical between the two.
Requires the ExecuTorch 1.4.1 Vulkan runtime from the v0.10.0-libs-1.4.1
artifacts.
