Crate spark_model 

Re-exports

pub use vision_item::VisionItem;

Modules

engine

Inference engine — generate loop for a single request.

factory

Model factory: builds the right model from config + weights.

forward

Vendor-agnostic per-layer forward orchestration.

kimi_k3

Kimi K3 host graph. This slice is KDA only.

layer

Composable transformer layer traits (SDD).

layers

lora

Startup-static PEFT LoRA adapter: remap/validate/pack into the fixed-address rank-padded pool. v0 = one adapter, slot 0, always on.

mistral_loader

Weight loader for Mistral Small 4 (MLA + MoE architecture).

model

Generic transformer model.

mtp_layout

Checkpoint-level detection of MTP / next-token-prediction weights.

precision_schedule

Per-layer + per-tensor precision overrides (C.3, 2026-04-25).

preflight

Pre-flight weight-store / config consistency checks.

quant_format

Weight-quantization format abstraction.

seq_state_reserve

Per-sequence device state the serve must RESERVE, computed before the model exists.

speculative

Speculative decoding abstraction (SDD).

ssm_reserve

SSOT for the SSM/linear-attention GPU reserve terms.

tp_shard

Tensor-parallel weight sharding helpers.

traits

Model trait (SDD: single trait, multiple implementations possible).

video_decode_ffmpeg

Frame extraction for real-world containers, via ffmpeg.

video_preprocess

Video → patch tensor, the temporal sibling of crate::vision_preprocess.

vision_item

The unit of vision input handed to the model.

vision_preprocess

CPU-side image preprocessing for Qwen3-VL vision inputs.

weight_loader

Weight loading traits and per-model loader implementations.

weight_map

Weight name mapping from HuggingFace safetensors to typed layer structures.

Constants

DEFAULT_MARCONI_MIN_TOKENS

Marconi snapshot-restore threshold: the shipped default, and the setter spark serve uses to pin it from --marconi-min-tokens before anything reads it. Re-exported rather than making model::mtp_carry public — the rest of that module is internal. The shipped threshold, and the only place it is written down.

Functions

model_type_ships_vanilla_norm_weights

The dispatch predicate itself, on the bare model_type, so it is unit-testable without constructing a full ModelConfig.

requires_single_chunk_prefill

Must chunked prefill run as a SINGLE chunk for this model?

set_marconi_min_tokens

Marconi snapshot-restore threshold: the shipped default, and the setter spark serve uses to pin it from --marconi-min-tokens before anything reads it. Re-exported rather than making model::mtp_carry public — the rest of that module is internal. Pin the restore threshold from --marconi-min-tokens, before anything reads it.

ships_vanilla_norm_weights

True when the checkpoint ships HF-vanilla RMSNorm weights — i.e. the norm weight is used as out = x * w / rms, not Qwen3-Next’s offset-from-1 out = x * (1 + w) / rms.