38 lines
1.1 KiB
YAML
38 lines
1.1 KiB
YAML
# Run with:
|
|
# FASTVIDEO_ATTENTION_BACKEND=VIDEO_SPARSE_ATTN fastvideo generate --config scripts/inference/inference_wan_VSA.yaml
|
|
generator:
|
|
model_path: FastVideo/Wan2.1-VSA-T2V-14B-720P-Diffusers
|
|
engine:
|
|
num_gpus: 1
|
|
parallelism:
|
|
tp_size: 1
|
|
sp_size: 1
|
|
offload:
|
|
dit: false
|
|
vae: false
|
|
pin_cpu_memory: false
|
|
pipeline:
|
|
experimental:
|
|
flow_shift: 5.0
|
|
VSA_sparsity: 0.9
|
|
request:
|
|
negative_prompt: >-
|
|
Bright tones, overexposed, static, blurred details, subtitles, style,
|
|
works, paintings, images, static, overall gray, worst quality, low
|
|
quality, JPEG compression residue, ugly, incomplete, extra fingers,
|
|
poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen
|
|
limbs, fused fingers, still picture, messy background, three legs, many
|
|
people in the background, walking backwards
|
|
inputs:
|
|
prompt_path: assets/prompt.txt
|
|
sampling:
|
|
seed: 1024
|
|
num_frames: 77
|
|
height: 448
|
|
width: 832
|
|
fps: 16
|
|
num_inference_steps: 50
|
|
guidance_scale: 5.0
|
|
output:
|
|
output_path: outputs_Wan-VSA-14B/
|