Skip to content

anyflow_far_transformer3d.md

Latest commit

 

History

History
45 lines (32 loc) · 2.16 KB

anyflow_far_transformer3d.md

File metadata and controls

45 lines (32 loc) · 2.16 KB

AnyFlowFARTransformer3DModel

The causal (FAR) 3D Transformer used by AnyFlowFARPipeline — the FAR variant of AnyFlow (Yuchao Gu, Guian Fang et al., NUS ShowLab × NVIDIA). It extends the v0.35.1 Wan2.1 backbone with three additions:

  1. FAR causal block-mask via torch.nn.attention.flex_attention, supporting frame-level autoregressive generation as introduced in FAR (Gu et al., 2025).
  2. Compressed-frame patch embedding (far_patch_embedding) for context (already-generated) frames, warm-started from the full-resolution patch_embedding at construction time via trilinear interpolation.
  3. Dual-timestep flow-map embedding (same as AnyFlowTransformer3DModel) — every forward call conditions on both the source timestep t and the target timestep r.

The chunk schedule (chunk_partition) is not baked into the model config. It is a per-call argument to forward, so the same checkpoint handles different num_frames configurations without retraining.

from diffusers import AnyFlowFARTransformer3DModel

# Causal AnyFlow checkpoint (FAR):
transformer = AnyFlowFARTransformer3DModel.from_pretrained(
    "nvidia/AnyFlow-FAR-Wan2.1-1.3B-Diffusers", subfolder="transformer"
)

AnyFlowFARTransformer3DModel

[[autodoc]] AnyFlowFARTransformer3DModel

AnyFlowFARTransformerOutput

[[autodoc]] models.transformers.transformer_anyflow_far.AnyFlowFARTransformerOutput