vllm_omni.diffusion.utils ¶

Modules:

Name	Description
`chunked_video`	Turn committed VAE chunks into MP4 bytes without materializing the video.
`flow_matching`
`future_utils`	Helpers for resolving `concurrent.futures.Future` objects shared across threads.
`hf_utils`
`image_output`
`kv_utils`	Utilities for batching variable-length tensors in diffusion attention.
`media_utils`	Video/audio muxing utilities using PyAV (no ffmpeg binary dependency).
`network_utils`
`param_utils`
`prompt_utils`
`size_utils`	Shared size normalization helpers for diffusion pipelines.
`tf_utils`
`video_decode`	Decode video file paths to RGB frames.