Expand description
2026-09-25: DFlash drafter loader: parses the drafter’s config.json and
loads its weights into DflashWeights for
crate::dflash_head::BlockDiffusionDraftHead.
The loader reads no embedding or LM head: BlockDiffusionDraftHead::from_weights
takes the target model’s. Tensor names may be bare or model.-prefixed.
Owner: model-arch weight loader. Invariants: none beyond the types.
Structs§
- Dflash
Config - 2026-09-25: The fields of a drafter’s
config.jsonthat the engine reads, under theirconfig.jsonnames. Unknown keys are ignored. - Dflash
Layer Weights - 2026-09-25: One drafter layer’s weights. The projections go through
dense_bf16_or_nvfp4; the norms and conv tensors are the store’s pointers. - Dflash
Rope Scaling - 2026-09-25: The
rope_scalingfields the drafter head reads. - Dflash
SubConfig - 2026-09-25: The nested
dflash_configblock of the drafter’sconfig.json. - Dflash
Weights - 2026-09-25: The drafter weights
load_dflash_weightsreturns. It holds no embedding or LM head.
Functions§
- load_
dflash_ weights - 2026-09-25: Load the drafter’s weights from the drafter’s own
WeightStore. - parse_
dflash_ config - 2026-09-25: Parse a drafter’s
config.jsoninto aDflashConfig. The error is “Parsing DFlash drafter config.json” with the serde error as its cause. - store_
has_ dflash_ weights - 2026-09-25: True when the store has
fc.weightormodel.fc.weight. Loads nothing.