presets

LingBotWorld2 causal-fast pipeline preset.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset

InferencePreset(name: str, version: int, model_family: str, description: str = '', workload_type: str | None = None, stage_schemas: tuple[PresetStageSpec, ...] = (), defaults: dict[str, Any] = dict(), stage_defaults: dict[str, dict[str, Any]] = dict())

A named inference preset for a model family.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.name

name: str

Preset name, e.g. "wan_t2v_1_3b".

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.version

version: int

Preset schema version; bump on breaking schema changes.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.model_family

model_family: str

Model family key, e.g. "wan", "ltx2".

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.workload_type

workload_type: str | None = None

Optional workload hint: "t2v", "i2v", etc.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.stage_schemas

stage_schemas: tuple[PresetStageSpec, ...] = ()

User-facing stage names for stage_overrides validation.

Validation-only: this tuple is consumed by :func:validate_stage_overrides and is not used to drive pipeline execution. Omit or leave empty if the preset exposes no per-stage override surface.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.defaults

defaults: dict[str, Any] = field(default_factory=dict)

Preset-level default sampling/runtime values.

fastvideo.pipelines.basic.lingbotworld2.presets.InferencePreset.stage_defaults

stage_defaults: dict[str, dict[str, Any]] = field(default_factory=dict)

Per-stage default overrides, keyed by stage name.

fastvideo.pipelines.basic.lingbotworld2.presets.PresetStageSpec

PresetStageSpec(name: str, kind: str, description: str = '', allowed_overrides: frozenset[str] = frozenset())

A user-facing stage name within a preset, used only to validate stage_overrides keys. Not read by pipeline execution — the real execution DAG lives in each pipeline's create_pipeline_stages().

fastvideo.pipelines.basic.lingbotworld2.presets.PresetStageSpec.name

name: str

Short user-facing name, e.g. "denoise", "sr".

fastvideo.pipelines.basic.lingbotworld2.presets.PresetStageSpec.kind

kind: str

Semantic kind, e.g. "denoising", "super_resolution".

fastvideo.pipelines.basic.lingbotworld2.presets.PresetStageSpec.allowed_overrides

allowed_overrides: frozenset[str] = field(default_factory=frozenset)

Keys that may appear in stage_overrides[name].