Expand description
Native Qwen4Exp asset preparation.
The v4 P64 format intentionally uses 32-bit relative offsets, so a single
P64 can never represent the 90+ GiB Qwen3.8 Flash Next checkpoint. The
correct native artefact today is a compact .hmc contract which attests to
the external GGUF and describes the PLE table’s direct NVMe byte span.
Future sub-4-GiB P64 trunk shards, .q42 policy volumes, and .10d
schedules are added as intact HMC entries without ever copying the PLE.
Structs§
- External
Gguf Asset - A verified external source. The HMC intentionally stores the identity and relative file name, not a host-specific absolute path.
- Qwen4
ExpNative Descriptor - Cold activation data placed in the
.hmcdescriptor entry. - Qwen4
ExpPle Span - Directly addressable PLE tensor span inside the external GGUF payload.
- Qwen4
ExpPle Storage - The extracted PLE payload. It is the only large native artifact: a raw, row-addressable 26.8 GiB table on the chosen NVMe volume, never a second GGUF or a RAM/VRAM-resident buffer.
Enums§
Constants§
Functions§
- load_
qwen4exp_ native_ hmc - Read the compact descriptor from a transparent HMC without loading any
external model bytes. The caller resolves and validates the external GGUF
with
Qwen4ExpNativeDescriptor::verify_source_metadataorQwen4ExpNativeDescriptor::verify_source_sha256. - prepare_
qwen4exp_ native_ hmc - Create a compact, immutable
.hmcactivation package on an artifact volume while its GGUF trunk remains in place. Only the PLE span is extracted tople_output; all other 60+ GiB stay in the source GGUF. An already-present PLE payload of the exact declared length is adopted without copying it again, allowing a compact HMC schema refresh without consuming disk space.