Anima per-step-expert turbo loader. Installs a diffusion_model pre-hook that selects the up-head by denoise-step counter (head k → step k) plus per-Linear…

Anima loader nodes for ComfyUI. AnimaAdapterLoader auto-detects LoRA, HydraLoRA multi-head with live σ-conditional and FeRA-style FEI-conditional routing, ChimeraHydra dual-pool routing (per-Linear content router + network-level FreqRouter on FEI+σ), and LoReFT residual-stream edits. AnimaFeraLoader handles FeRA (Yin et al., arXiv:2511.17979) — both the author-faithful networks.methods.fera format and the plan2 stacked_experts_global_fei format with global router on the latent's spectral energy + independent stacked experts per Linear. AnimaSoftTokensLoader handles SoftREPA-parameterization soft tokens (Lee et al., arXiv:2503.08250) — per-layer × per-timestep-bucket learned tokens spliced into the crossattn embedding inside the first n_layers DiT blocks. Three single-purpose nodes; chain via the MODEL socket.

Anima Adapter Loader

model

adapter

strength

Anima per-step-expert turbo loader. Installs a diffusion_model pre-hook that selects the up-head by denoise-step counter (head k → step k) plus per-Linear forward hooks. Fused qkv/kv keys are split to ComfyUI's q/k/v layout at load. Requires cfg=1.0 and one model forward per step (the turbo contract); set infer_steps to the trained K.

Name	Type	Default	Description
model	MODEL		—
adapter	COMBO		Per-step-expert turbo student (ss_turbo_per_step_expert=1). Use cfg=1.0 and infer_steps = trained head count K.
strength	FLOAT	1.00-2–2	Strength multiplier for the step-head delta.

Anima Turbo Per-Step Expert Loader

Inputs (3)

Outputs (1)