mindone.diffusers is compatible with 🤗 diffusers v0.35.2, and provides preview supports for SoTA v0.36 pipelines. Please refer to the docs for detailed limitations and usage of each component.
- 240+ diffusion pipelines for inference
- 70+ pretrained models (AutoEncoder, ControlNet, Transformer, UNet) for constructing pipelines
- 40+ noise schedulers to control diffusion speed and quality
Support list for pipelines
- fp32/fp16/bf16: ✅ = passed fast UT for that precision (performed on pruned models)
- inference: ✅ = verified with official released weights
- ✖️ = not verified yet or temporarily OOM
| pipelines | fp32 | fp16 | bf16 | inference |
|---|---|---|---|---|
| allegro | ✖️ | ✅ | ✅ | ✅ |
| amused | ✅ | ✅ | ✖️ | ✅ |
| amused_img2img | ✅ | ✅ | ✖️ | ✅ |
| amused_inpaint | ✅ | ✅ | ✖️ | ✅ |
| animatediff | ✅ | ✅ | ✖️ | ✅ |
| animatediff_controlnet | ✅ | ✅ | ✖️ | ✅ |
| animatediff_sdxl | ✅ | ✅ | ✖️ | ✅ |
| animatediff_sparsectrl | ✅ | ✅ | ✖️ | ✅ |
| animatediff_video2video | ✅ | ✅ | ✖️ | ✅ |
| audioldm | ✅ | ✅ | ✖️ | ✅ |
| audioldm2 | ✅ | ✅ | ✖️ | ✅ |
| aura_flow | ✅ | ✅ | ✖️ | ✅ |
| blip_diffusion | ✅ | ✅ | ✖️ | ✅ |
| bria | ✖️ | ✅ | ✅ | ✅ |
| chroma | ✅ | ✅ | ✖️ | ✖️ |
| chroma_img2img | ✅ | ✅ | ✖️ | ✖️ |
| cogview3plus | ✅ | ✖️ | ✅ | ✅ |
| cogview4 | ✖️ | ✖️ | ✖️ | ✖️ |
| cogview4_control | ✖️ | ✖️ | ✖️ | ✖️ |
| cogvideox | ✖️ | ✅ | ✅ | ✅ |
| cogvideox_fun_control | ✖️ | ✅ | ✅ | ✅ |
| cogvideox_image2video | ✖️ | ✅ | ✅ | ✅ |
| cogvideox_video2video | ✖️ | ✅ | ✅ | ✅ |
| consisid | ✖️ | ✅ | ✅ | ✖️ |
| consistency_models | ✅ | ✅ | ✖️ | ✅ |
| controlnet | ✅ | ✅ | ✖️ | ✅ |
| controlnet_blip_diffusion | ✅ | ✅ | ✖️ | ✅ |
| controlnet_img2img | ✅ | ✅ | ✖️ | ✅ |
| controlnet_inpaint | ✅ | ✅ | ✖️ | ✅ |
| controlnet_inpaint_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| controlnet_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| controlnet_sd_xl_img2img | ✅ | ✅ | ✖️ | ✅ |
| controlnet_union_inpaint_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| controlnet_union_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| controlnet_union_sd_xl_img2img | ✅ | ✅ | ✖️ | ✅ |
| controlnet_xs | ✅ | ✅ | ✖️ | ✅ |
| controlnet_xs_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| cosmos2_text2image | ✖️ | ✅ | ✅ | ✅ |
| cosmos2_video2world | ✖️ | ✅ | ✅ | ✅ |
| cosmos_text2world | ✖️ | ✅ | ✅ | ✖️ |
| cosmos_video2world | ✖️ | ✅ | ✅ | ✅ |
| dance_diffusion | ✅ | ✅ | ✖️ | ✅ |
| ddim | ✅ | ✅ | ✖️ | ✅ |
| ddpm | ✅ | ✅ | ✖️ | ✅ |
| dit | ✅ | ✅ | ✖️ | ✅ |
| easyanimate | ✖️ | ✖️ | ✖️ | ✖️ |
| easyanimate_control | ✖️ | ✖️ | ✖️ | ✖️ |
| easyanimate_inpaint | ✖️ | ✖️ | ✖️ | ✖️ |
| flux | ✅ | ✅ | ✖️ | ✅ |
| flux_control | ✖️ | ✅ | ✅ | ✅ |
| flux_control_img2img | ✖️ | ✅ | ✅ | ✅ |
| flux_control_inpaint | ✖️ | ✅ | ✅ | ✅ |
| flux_controlnet | ✖️ | ✅ | ✅ | ✅ |
| flux_controlnet_image_to_image | ✅ | ✅ | ✖️ | ✅ |
| flux_controlnet_inpainting | ✅ | ✅ | ✖️ | ✅ |
| flux_fill | ✖️ | ✅ | ✅ | ✅ |
| flux_img2img | ✖️ | ✅ | ✅ | ✅ |
| flux_inpaint | ✖️ | ✅ | ✅ | ✅ |
| flux_kontext | ✖️ | ✖️ | ✅ | ✅ |
| flux_kontext_inpaint | ✖️ | ✅ | ✅ | ✅ |
| flux_prior_redux | ✖️ | ✅ | ✅ | ✅ |
| hidream_image | ✅ | ✅ | ✅ | ✖️ |
| hunyuan_skyreels_image2video | ✖️ | ✖️ | ✖️ | ✖️ |
| hunyuan_video | ✅ | ✅ | ✖️ | ✅ |
| hunyuan_video_framepack | ✅ | ✅ | ✅ | ✅ |
| hunyuan_video_image2video | ✖️ | ✖️ | ✖️ | ✖️ |
| hunyuandit | ✅ | ✅ | ✖️ | ✅ |
| hunyuandit_controlnet | ✅ | ✅ | ✖️ | ✅ |
| i2vgen_xl | ✖️ | ✅ | ✖️ | ✅ |
| if | ✅ | ✅ | ✖️ | ✅ |
| if_img2img | ✅ | ✅ | ✖️ | ✅ |
| if_img2img_superresolution | ✅ | ✅ | ✖️ | ✅ |
| if_inpainting | ✅ | ✅ | ✖️ | ✅ |
| if_inpainting_superresolution | ✅ | ✅ | ✖️ | ✅ |
| if_superresolution | ✅ | ✅ | ✖️ | ✅ |
| kandinsky | ✅ | ✅ | ✖️ | ✅ |
| kandinsky_combined | ✅ | ✅ | ✖️ | ✅ |
| kandinsky_img2img | ✅ | ✅ | ✖️ | ✅ |
| kandinsky_inpaint | ✅ | ✅ | ✖️ | ✅ |
| kandinsky_prior | ✅ | ✅ | ✖️ | ✖️ |
| kandinsky2_2 | ✅ | ✅ | ✖️ | ✅ |
| kandinsky2_2_combined | ✅ | ✅ | ✖️ | ✅ |
| kandinsky2_2_controlnet | ✅ | ✅ | ✖️ | ✅ |
| kandinsky2_2_controlnet_img2img | ✅ | ✅ | ✖️ | ✅ |
| kandinsky2_2_img2img | ✅ | ✅ | ✖️ | ✅ |
| kandinsky2_2_inpainting | ✖️ | ✖️ | ✖️ | ✖️ |
| kandinsky2_2_prior | ✅ | ✅ | ✖️ | ✖️ |
| kandinsky2_2_prior_emb2emb | ✅ | ✅ | ✖️ | ✖️ |
| kandinsky3 | ✅ | ✅ | ✖️ | ✅ |
| kandinsky3_img2img | ✅ | ✅ | ✖️ | ✅ |
| kolors | ✅ | ✅ | ✖️ | ✅ |
| kolors_img2img | ✅ | ✅ | ✖️ | ✅ |
| latent_consistency_img2img | ✅ | ✅ | ✖️ | ✅ |
| latent_consistency_text2img | ✅ | ✅ | ✖️ | ✅ |
| latent_diffusion | ✅ | ✅ | ✖️ | ✅ |
| latent_diffusion_superresolution | ✅ | ✅ | ✖️ | ✅ |
| latte | ✅ | ✅ | ✖️ | ✅ |
| leditspp_stable_diffusion | ✅ | ✅ | ✖️ | ✅ |
| leditspp_stable_diffusion_xl | ✅ | ✅ | ✖️ | ✅ |
| ltx | ✅ | ✅ | ✅ | ✅ |
| ltx_condition | ✅ | ✅ | ✖️ | ✅ |
| ltx_image2video | ✅ | ✅ | ✖️ | ✅ |
| ltx_latent_upsample | ✅ | ✅ | ✅ | ✅ |
| lucy_edit | ✖️ | ✖️ | ✖️ | ✖️ |
| lumina | ✅ | ✖️ | ✅ | ✅ |
| lumina2 | ✅ | ✖️ | ✅ | ✅ |
| marigold_depth | ✅ | ✅ | ✖️ | ✅ |
| marigold_intrinsics | ✅ | ✅ | ✖️ | ✅ |
| marigold_normals | ✅ | ✅ | ✖️ | ✅ |
| mochi | ✖️ | ✅ | ✖️ | ✅ |
| musicldm | ✅ | ✅ | ✖️ | ✅ |
| omnigen | ✅ | ✅ | ✅ | ✅ |
| pag_controlnet_sd | ✅ | ✅ | ✖️ | ✅ |
| pag_controlnet_sd_inpaint | ✅ | ✅ | ✖️ | ✅ |
| pag_controlnet_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| pag_controlnet_sd_xl_img2img | ✅ | ✅ | ✖️ | ✅ |
| pag_hunyuandit | ✅ | ✅ | ✖️ | ✅ |
| pag_kolors | ✅ | ✅ | ✖️ | ✅ |
| pag_pixart_sigma | ✅ | ✅ | ✖️ | ✅ |
| pag_sana | ✅ | ✅ | ✅ | ✅ |
| pag_sd | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_3 | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_3_img2img | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_animatediff | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_img2img | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_inpaint | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_xl | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_xl_img2img | ✅ | ✅ | ✖️ | ✅ |
| pag_sd_xl_inpaint | ✅ | ✅ | ✖️ | ✅ |
| paint_by_example | ✅ | ✅ | ✖️ | ✅ |
| pia | ✅ | ✅ | ✖️ | ✅ |
| pixart_alpha | ✅ | ✅ | ✖️ | ✅ |
| pixart_sigma | ✅ | ✅ | ✖️ | ✅ |
| qwenimage | ✅ | ✖️ | ✅ | ✅ |
| qwenimage_edit | ✅ | ✖️ | ✅ | ✅ |
| qwenimage_edit_inpaint | ✖️ | ✖️ | ✖️ | ✖️ |
| qwenimage_img2img | ✅ | ✖️ | ✅ | ✅ |
| qwenimage_inpaint | ✅ | ✖️ | ✅ | ✅ |
| sana | ✅ | ✅ | ✖️ | ✅ |
| sana_controlnet | ✅ | ✅ | ✅ | ✅ |
| sana_sprint | ✅ | ✖️ | ✅ | ✅ |
| sana_sprint_img2img | ✅ | ✅ | ✅ | ✅ |
| semantic_stable_diffusion | ✅ | ✅ | ✖️ | ✅ |
| shap_e | ✅ | ✅ | ✖️ | ✅ |
| shap_e_img2img | ✅ | ✅ | ✖️ | ✅ |
| skyreels_v2 | ✅ | ✅ | ✅ | ✅ |
| skyreels_v2_diffusion_forcing | ✅ | ✅ | ✅ | ✅ |
| skyreels_v2_diffusion_forcing_i2v | ✅ | ✅ | ✅ | ✅ |
| skyreels_v2_diffusion_forcing_v2v | ✅ | ✅ | ✅ | ✅ |
| skyreels_v2_i2v | ✅ | ✅ | ✅ | ✅ |
| stable_audio | ✅ | ✖️ | ✖️ | ✅ |
| stable_cascade | ✅ | ✅ | ✖️ | ✅ |
| stable_cascade_combined | ✅ | ✅ | ✖️ | ✅ |
| stable_cascade_prior | ✅ | ✅ | ✖️ | ✖️ |
| stable_diffusion | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_3 | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_3_controlnet | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_3_controlnet_inpainting | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_3_img2img | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_3_inpaint | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_adapter | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_attend_and_excite | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_depth2img | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_diffedit | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_gligen | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_gligen_text_image | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_image_variation | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_img2img | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_inpaint | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_instruct_pix2pix | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_k_diffusion | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_latent_upscale | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_ldm3d | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_panorama | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_safe | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_sag | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_upscale | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl_adapter | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl_img2img | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl_inpaint | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl_instruct_pix2pix | ✅ | ✅ | ✖️ | ✅ |
| stable_diffusion_xl_k_diffusion | ✅ | ✅ | ✖️ | ✅ |
| stable_unclip | ✅ | ✅ | ✖️ | ✅ |
| stable_unclip_img2img | ✅ | ✅ | ✖️ | ✅ |
| stable_video_diffusion | ✖️ | ✅ | ✖️ | ✅ |
| text_to_video_synth | ✅ | ✅ | ✖️ | ✅ |
| text_to_video_synth_img2img | ✅ | ✅ | ✖️ | ✅ |
| text_to_video_zero | ✅ | ✅ | ✖️ | ✅ |
| text_to_video_zero_sdxl | ✅ | ✅ | ✖️ | ✅ |
| unclip | ✅ | ✅ | ✖️ | ✅ |
| unclip_image_variation | ✅ | ✅ | ✖️ | ✅ |
| unidiffuser | ✅ | ✅ | ✖️ | ✅ |
| visualcloze_combined | ✅ | ✅ | ✅ | ✅ |
| visualcloze_generation | ✅ | ✅ | ✅ | ✅ |
| wan | ✅ | ✅ | ✅ | ✅ |
| wan_i2v | ✅ | ✅ | ✅ | ✅ |
| wan_vace | ✅ | ✅ | ✅ | ✅ |
| wan_video2video | ✅ | ✅ | ✅ | ✅ |
| wuerstchen | ✅ | ✅ | ✖️ | ✅ |
| wuerstchen_combined | ✅ | ✅ | ✖️ | ✅ |
| wuerstchen_prior | ✅ | ✅ | ✖️ | ✖️ |