mirror of
https://github.com/modelscope/DiffSynth-Studio.git
synced 2026-03-19 14:58:12 +00:00
2.6 KiB
2.6 KiB
Pipelines
DiffSynth-Studio includes multiple pipelines, categorized into two types: image generation and video generation.
Image Pipelines
| Pipeline | Models |
|---|---|
| SDImagePipeline | text_encoder: SDTextEncoder unet: SDUNet vae_decoder: SDVAEDecoder vae_encoder: SDVAEEncoder controlnet: MultiControlNetManager ipadapter_image_encoder: IpAdapterCLIPImageEmbedder ipadapter: SDIpAdapter |
| SDXLImagePipeline | text_encoder: SDXLTextEncoder text_encoder_2: SDXLTextEncoder2 text_encoder_kolors: ChatGLMModel unet: SDXLUNet vae_decoder: SDXLVAEDecoder vae_encoder: SDXLVAEEncoder controlnet: MultiControlNetManager ipadapter_image_encoder: IpAdapterXLCLIPImageEmbedder ipadapter: SDXLIpAdapter |
| SD3ImagePipeline | text_encoder_1: SD3TextEncoder1 text_encoder_2: SD3TextEncoder2 text_encoder_3: SD3TextEncoder3 dit: SD3DiT vae_decoder: SD3VAEDecoder vae_encoder: SD3VAEEncoder |
| HunyuanDiTImagePipeline | text_encoder: HunyuanDiTCLIPTextEncoder text_encoder_t5: HunyuanDiTT5TextEncoder dit: HunyuanDiT vae_decoder: SDVAEDecoder vae_encoder: SDVAEEncoder |
| FluxImagePipeline | text_encoder_1: FluxTextEncoder1 text_encoder_2: FluxTextEncoder2 dit: FluxDiT vae_decoder: FluxVAEDecoder vae_encoder: FluxVAEEncoder |
Video Pipelines
| Pipeline | Models |
|---|---|
| SDVideoPipeline | text_encoder: SDTextEncoder unet: SDUNet vae_decoder: SDVAEDecoder vae_encoder: SDVAEEncoder controlnet: MultiControlNetManager ipadapter_image_encoder: IpAdapterCLIPImageEmbedder ipadapter: SDIpAdapter motion_modules: SDMotionModel |
| SDXLVideoPipeline | text_encoder: SDXLTextEncoder text_encoder_2: SDXLTextEncoder2 text_encoder_kolors: ChatGLMModel unet: SDXLUNet vae_decoder: SDXLVAEDecoder vae_encoder: SDXLVAEEncoder ipadapter_image_encoder: IpAdapterXLCLIPImageEmbedder ipadapter: SDXLIpAdapter motion_modules: SDXLMotionModel |
| SVDVideoPipeline | image_encoder: SVDImageEncoder unet: SVDUNet vae_encoder: SVDVAEEncoder vae_decoder: SVDVAEDecoder |
| CogVideoPipeline | text_encoder: FluxTextEncoder2 dit: CogDiT vae_encoder: CogVAEEncoder vae_decoder: CogVAEDecoder |