theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-04-24 15:06:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

sd

Browse Source
This commit is contained in:
mi804
2026-04-23 17:35:24 +08:00
parent 5c89a15b9a
commit 82e482286c
14 changed files with 3274 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
diffsynth/configs/model_configs.py
View File

@@ -900,6 +900,53 @@ mova_series = [
"model_class": "diffsynth.models.mova_dual_tower_bridge.DualTowerConditionalBridge",
},
]
stable_diffusion_xl_series = [
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-xl-base-1.0", origin_file_pattern="unet/diffusion_pytorch_model.safetensors")
"model_hash": "142b114f67f5ab3a6d83fb5788f12ded",
"model_name": "stable_diffusion_xl_unet",
"model_class": "diffsynth.models.stable_diffusion_xl_unet.SDXLUNet2DConditionModel",
},
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-xl-base-1.0", origin_file_pattern="text_encoder_2/model.safetensors")
"model_hash": "98cc34ccc5b54ae0e56bdea8688dcd5a",
"model_name": "stable_diffusion_xl_text_encoder",
"model_class": "diffsynth.models.stable_diffusion_xl_text_encoder.SDXLTextEncoder2",
"state_dict_converter": "diffsynth.utils.state_dict_converters.stable_diffusion_xl_text_encoder.SDXLTextEncoder2StateDictConverter",
},
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-xl-base-1.0", origin_file_pattern="vae/diffusion_pytorch_model.safetensors")
"model_hash": "13115dd45a6e1c39860f91ab073b8a78",
"model_name": "stable_diffusion_xl_vae",
"model_class": "diffsynth.models.stable_diffusion_vae.StableDiffusionVAE",
"state_dict_converter": "diffsynth.utils.state_dict_converters.stable_diffusion_vae.SDVAEStateDictConverter",
"extra_kwargs": {"scaling_factor": 0.13025, "sample_size": 1024, "force_upcast": True},
},
]
stable_diffusion_series = [
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-v1-5", origin_file_pattern="text_encoder/model.safetensors")
"model_hash": "ffd1737ae9df7fd43f5fbed653bdad67",
"model_name": "stable_diffusion_text_encoder",
"model_class": "diffsynth.models.stable_diffusion_text_encoder.SDTextEncoder",
"state_dict_converter": "diffsynth.utils.state_dict_converters.stable_diffusion_text_encoder.SDTextEncoderStateDictConverter",
},
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-v1-5", origin_file_pattern="vae/diffusion_pytorch_model.safetensors")
"model_hash": "f86d5683ed32433be8ca69969c67ba69",
"model_name": "stable_diffusion_vae",
"model_class": "diffsynth.models.stable_diffusion_vae.StableDiffusionVAE",
"state_dict_converter": "diffsynth.utils.state_dict_converters.stable_diffusion_vae.SDVAEStateDictConverter",
},
{
# Example: ModelConfig(model_id="AI-ModelScope/stable-diffusion-v1-5", origin_file_pattern="unet/diffusion_pytorch_model.safetensors")
"model_hash": "025a4b86a84829399d89f613e580757b",
"model_name": "stable_diffusion_unet",
"model_class": "diffsynth.models.stable_diffusion_unet.UNet2DConditionModel",
},
]
joyai_image_series = [
{
# Example: ModelConfig(model_id="jd-opensource/JoyAI-Image-Edit", origin_file_pattern="transformer/transformer.pth")
@@ -916,4 +963,4 @@ joyai_image_series = [
},
]
MODEL_CONFIGS = qwen_image_series + wan_series + flux_series + flux2_series + ernie_image_series + z_image_series + ltx2_series + anima_series + mova_series + joyai_image_series
MODEL_CONFIGS = qwen_image_series + wan_series + flux_series + flux2_series + ernie_image_series + z_image_series + ltx2_series + anima_series + mova_series + stable_diffusion_xl_series + stable_diffusion_series + joyai_image_series

39
diffsynth/configs/vram_management_module_maps.py
View File

@@ -295,6 +295,45 @@ VRAM_MANAGEMENT_MODULE_MAPS = {
"transformers.models.qwen3_vl.modeling_qwen3_vl.Qwen3VLTextRMSNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.qwen3_vl.modeling_qwen3_vl.Qwen3VLTextRotaryEmbedding": "diffsynth.core.vram.layers.AutoWrappedModule",
},
"diffsynth.models.stable_diffusion_unet.UNet2DConditionModel": {
"torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
"torch.nn.Conv2d": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.GroupNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.LayerNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.SiLU": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.Dropout": "diffsynth.core.vram.layers.AutoWrappedModule",
},
"diffsynth.models.stable_diffusion_vae.StableDiffusionVAE": {
"torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
"torch.nn.Conv2d": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.GroupNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.SiLU": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.Dropout": "diffsynth.core.vram.layers.AutoWrappedModule",
"diffsynth.models.stable_diffusion_vae.Upsample2D": "diffsynth.core.vram.layers.AutoWrappedModule",
"diffsynth.models.stable_diffusion_vae.Downsample2D": "diffsynth.core.vram.layers.AutoWrappedModule",
},
"diffsynth.models.stable_diffusion_text_encoder.SDTextEncoder": {
"torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
"torch.nn.Embedding": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPTextTransformer": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPEncoderLayer": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPAttention": "diffsynth.core.vram.layers.AutoWrappedModule",
},
"diffsynth.models.stable_diffusion_xl_unet.SDXLUNet2DConditionModel": {
"torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
"torch.nn.Conv2d": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.GroupNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.LayerNorm": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.SiLU": "diffsynth.core.vram.layers.AutoWrappedModule",
"torch.nn.Dropout": "diffsynth.core.vram.layers.AutoWrappedModule",
},
"diffsynth.models.stable_diffusion_xl_text_encoder.SDXLTextEncoder2": {
"torch.nn.Linear": "diffsynth.core.vram.layers.AutoWrappedLinear",
"torch.nn.Embedding": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPTextTransformer": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPEncoderLayer": "diffsynth.core.vram.layers.AutoWrappedModule",
"transformers.models.clip.modeling_clip.CLIPAttention": "diffsynth.core.vram.layers.AutoWrappedModule",
},
}
def QwenImageTextEncoder_Module_Map_Updater():