fastvideo.configs.sample.wan

Contents

`fastvideo.configs.sample.wan`#

Module Contents#

Classes#

`FastWanT2V480PConfig`
`Wan2_2_Base_SamplingParam`	Sampling parameters for Wan2.2 TI2V 5B model.
`Wan2_2_I2V_A14B_SamplingParam`
`Wan2_2_T2V_A14B_SamplingParam`
`Wan2_2_TI2V_5B_SamplingParam`	Sampling parameters for Wan2.2 TI2V 5B model.
`WanI2V_14B_480P_SamplingParam`
`WanI2V_14B_720P_SamplingParam`
`WanT2V_14B_SamplingParam`
`WanT2V_1_3B_SamplingParam`

API#

class fastvideo.configs.sample.wan.FastWanT2V480PConfig[source]#

Bases: fastvideo.configs.sample.wan.WanT2V_1_3B_SamplingParam

fps: int[source]#: 16

height: int[source]#: 448

num_frames: int[source]#: 61

num_inference_steps: int[source]#: 3

width: int[source]#: 832

class fastvideo.configs.sample.wan.Wan2_2_Base_SamplingParam[source]#

Bases: fastvideo.configs.sample.base.SamplingParam

Sampling parameters for Wan2.2 TI2V 5B model.

negative_prompt: str | None[source]#: ‘色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸…’

class fastvideo.configs.sample.wan.Wan2_2_I2V_A14B_SamplingParam[source]#: Bases: fastvideo.configs.sample.wan.Wan2_2_Base_SamplingParam

class fastvideo.configs.sample.wan.Wan2_2_T2V_A14B_SamplingParam[source]#: Bases: fastvideo.configs.sample.wan.Wan2_2_Base_SamplingParam

class fastvideo.configs.sample.wan.Wan2_2_TI2V_5B_SamplingParam[source]#

Bases: fastvideo.configs.sample.wan.Wan2_2_Base_SamplingParam

Sampling parameters for Wan2.2 TI2V 5B model.

fps: int[source]#: 24

guidance_scale: float[source]#: 5.0

height: int[source]#: 704

num_frames: int[source]#: 121

num_inference_steps: int[source]#: 50

width: int[source]#: 1280

class fastvideo.configs.sample.wan.WanI2V_14B_480P_SamplingParam[source]#

Bases: fastvideo.configs.sample.wan.WanT2V_1_3B_SamplingParam

guidance_scale: float[source]#: 5.0

num_inference_steps: int[source]#: 40

teacache_params: fastvideo.configs.sample.teacache.WanTeaCacheParams[source]#: ‘field(…)’

class fastvideo.configs.sample.wan.WanI2V_14B_720P_SamplingParam[source]#

Bases: fastvideo.configs.sample.wan.WanT2V_14B_SamplingParam

guidance_scale: float[source]#: 5.0

num_inference_steps: int[source]#: 40

teacache_params: fastvideo.configs.sample.teacache.WanTeaCacheParams[source]#: ‘field(…)’

class fastvideo.configs.sample.wan.WanT2V_14B_SamplingParam[source]#

Bases: fastvideo.configs.sample.base.SamplingParam

fps: int[source]#: 16

guidance_scale: float[source]#: 5.0

height: int[source]#: 720

negative_prompt: str[source]#: ‘Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, stat…’

num_frames: int[source]#: 81

num_inference_steps: int[source]#: 50

teacache_params: fastvideo.configs.sample.teacache.WanTeaCacheParams[source]#: ‘field(…)’

width: int[source]#: 1280

class fastvideo.configs.sample.wan.WanT2V_1_3B_SamplingParam[source]#

Bases: fastvideo.configs.sample.base.SamplingParam

fps: int[source]#: 16

guidance_scale: float[source]#: 3.0

height: int[source]#: 480

negative_prompt: str[source]#: ‘Bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, stat…’

num_frames: int[source]#: 81

num_inference_steps: int[source]#: 50

teacache_params: fastvideo.configs.sample.teacache.WanTeaCacheParams[source]#: ‘field(…)’

width: int[source]#: 832