Upload 19 files

Zero123++ model upload

README.md

@@ -0,0 +1,35 @@
+---
+license: openrail
+datasets:
+- allenai/objaverse
+library_name: diffusers
+pipeline_tag: image-to-image
+tags:
+- art
+---
+
+Recommended version of `diffusers` is `0.20.2` with `torch` `2`.
+
+Usage Example:
+```python
+import torch
+import requests
+from PIL import Image
+from diffusers import DiffusionPipeline, EulerAncestralDiscreteScheduler
+
+# Load the pipeline
+pipeline = DiffusionPipeline.from_pretrained(
+    "sudo-ai/zero123plus-v1.1", custom_pipeline="sudo-ai/zero123plus-pipeline",
+    torch_dtype=torch.float16
+)
+# Feel free to tune the scheduler
+pipeline.scheduler = EulerAncestralDiscreteScheduler.from_config(
+    pipeline.scheduler.config, timestep_spacing='trailing'
+)
+pipeline.to('cuda:0')
+# Run the pipeline
+cond = Image.open(requests.get("https://d.skis.ltd/nrp/sample-data/lysol.png", stream=True).raw)
+result = pipeline(cond).images[0]
+result.show()
+result.save("output.png")
+```

feature_extractor_clip/preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 224
+  }
+}

feature_extractor_vae/preprocessor_config.json

@@ -0,0 +1,19 @@
+{
+  "crop_size": {
+    "height": 512,
+    "width": 512
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": 0.5,
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": 0.8,
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 512
+  }
+}

inference.py

@@ -0,0 +1,390 @@
+from typing import Any, Dict, Optional
+from diffusers.models import AutoencoderKL, UNet2DConditionModel
+from diffusers.schedulers import KarrasDiffusionSchedulers
+
+import numpy
+import torch
+import torch.nn as nn
+import torch.utils.checkpoint
+import torch.distributed
+import transformers
+from collections import OrderedDict
+from PIL import Image
+from torchvision import transforms
+from transformers import CLIPImageProcessor, CLIPTextModel, CLIPTokenizer
+
+import diffusers
+from diffusers import (
+    AutoencoderKL,
+    DDPMScheduler,
+    DiffusionPipeline,
+    EulerAncestralDiscreteScheduler,
+    UNet2DConditionModel,
+    ImagePipelineOutput
+)
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.models.attention_processor import Attention, AttnProcessor, XFormersAttnProcessor, AttnProcessor2_0
+from diffusers.utils.import_utils import is_xformers_available
+
+
+def to_rgb_image(maybe_rgba: Image.Image):
+    if maybe_rgba.mode == 'RGB':
+        return maybe_rgba
+    elif maybe_rgba.mode == 'RGBA':
+        rgba = maybe_rgba
+        img = numpy.random.randint(127, 128, size=[rgba.size[1], rgba.size[0], 3], dtype=numpy.uint8)
+        img = Image.fromarray(img, 'RGB')
+        img.paste(rgba, mask=rgba.getchannel('A'))
+        return img
+    else:
+        raise ValueError("Unsupported image type.", maybe_rgba.mode)
+
+
+class ReferenceOnlyAttnProc(torch.nn.Module):
+    def __init__(
+        self,
+        chained_proc,
+        enabled=False,
+        name=None
+    ) -> None:
+        super().__init__()
+        self.enabled = enabled
+        self.chained_proc = chained_proc
+        self.name = name
+
+    def __call__(
+        self, attn: Attention, hidden_states, encoder_hidden_states=None, attention_mask=None,
+        mode="w", ref_dict: dict = None, is_cfg_guidance = False
+    ) -> Any:
+        if encoder_hidden_states is None:
+            encoder_hidden_states = hidden_states
+        if self.enabled and is_cfg_guidance:
+            res0 = self.chained_proc(attn, hidden_states[:1], encoder_hidden_states[:1], attention_mask)
+            hidden_states = hidden_states[1:]
+            encoder_hidden_states = encoder_hidden_states[1:]
+        if self.enabled:
+            if mode == 'w':
+                ref_dict[self.name] = encoder_hidden_states
+            elif mode == 'r':
+                encoder_hidden_states = torch.cat([encoder_hidden_states, ref_dict.pop(self.name)], dim=1)
+            elif mode == 'm':
+                encoder_hidden_states = torch.cat([encoder_hidden_states, ref_dict[self.name]], dim=1)
+            else:
+                assert False, mode
+        res = self.chained_proc(attn, hidden_states, encoder_hidden_states, attention_mask)
+        if self.enabled and is_cfg_guidance:
+            res = torch.cat([res0, res])
+        return res
+
+
+class RefOnlyNoisedUNet(torch.nn.Module):
+    def __init__(self, unet: UNet2DConditionModel, train_sched: DDPMScheduler, val_sched: EulerAncestralDiscreteScheduler) -> None:
+        super().__init__()
+        self.unet = unet
+        self.train_sched = train_sched
+        self.val_sched = val_sched
+
+        unet_lora_attn_procs = dict()
+        for name, _ in unet.attn_processors.items():
+            if torch.__version__ >= '2.0':
+                default_attn_proc = AttnProcessor2_0()
+            elif is_xformers_available():
+                default_attn_proc = XFormersAttnProcessor()
+            else:
+                default_attn_proc = AttnProcessor()
+            unet_lora_attn_procs[name] = ReferenceOnlyAttnProc(
+                default_attn_proc, enabled=name.endswith("attn1.processor"), name=name
+            )
+        unet.set_attn_processor(unet_lora_attn_procs)
+
+    def __getattr__(self, name: str):
+        try:
+            return super().__getattr__(name)
+        except AttributeError:
+            return getattr(self.unet, name)
+
+    def forward_cond(self, noisy_cond_lat, timestep, encoder_hidden_states, class_labels, ref_dict, is_cfg_guidance, **kwargs):
+        if is_cfg_guidance:
+            encoder_hidden_states = encoder_hidden_states[1:]
+            class_labels = class_labels[1:]
+        self.unet(
+            noisy_cond_lat, timestep,
+            encoder_hidden_states=encoder_hidden_states,
+            class_labels=class_labels,
+            cross_attention_kwargs=dict(mode="w", ref_dict=ref_dict),
+            **kwargs
+        )
+
+    def forward(
+        self, sample, timestep, encoder_hidden_states, class_labels=None,
+        *args, cross_attention_kwargs,
+        down_block_res_samples=None, mid_block_res_sample=None,
+        **kwargs
+    ):
+        cond_lat = cross_attention_kwargs['cond_lat']
+        is_cfg_guidance = cross_attention_kwargs.get('is_cfg_guidance', False)
+        noise = torch.randn_like(cond_lat)
+        if self.training:
+            noisy_cond_lat = self.train_sched.add_noise(cond_lat, noise, timestep)
+            noisy_cond_lat = self.train_sched.scale_model_input(noisy_cond_lat, timestep)
+        else:
+            noisy_cond_lat = self.val_sched.add_noise(cond_lat, noise, timestep.reshape(-1))
+            noisy_cond_lat = self.val_sched.scale_model_input(noisy_cond_lat, timestep.reshape(-1))
+        ref_dict = {}
+        self.forward_cond(
+            noisy_cond_lat, timestep,
+            encoder_hidden_states, class_labels,
+            ref_dict, is_cfg_guidance, **kwargs
+        )
+        weight_dtype = self.unet.dtype
+        return self.unet(
+            sample, timestep,
+            encoder_hidden_states, *args,
+            class_labels=class_labels,
+            cross_attention_kwargs=dict(mode="r", ref_dict=ref_dict, is_cfg_guidance=is_cfg_guidance),
+            down_block_additional_residuals=[
+                sample.to(dtype=weight_dtype) for sample in down_block_res_samples
+            ] if down_block_res_samples is not None else None,
+            mid_block_additional_residual=(
+                mid_block_res_sample.to(dtype=weight_dtype)
+                if mid_block_res_sample is not None else None
+            ),
+            **kwargs
+        )
+
+
+def scale_latents(latents):
+    latents = (latents - 0.22) * 0.75
+    return latents
+
+
+def unscale_latents(latents):
+    latents = latents / 0.75 + 0.22
+    return latents
+
+
+def scale_image(image):
+    image = image * 0.5 / 0.8
+    return image
+
+
+def unscale_image(image):
+    image = image / 0.5 * 0.8
+    return image
+
+
+class DepthControlUNet(torch.nn.Module):
+    def __init__(self, unet: RefOnlyNoisedUNet) -> None:
+        super().__init__()
+        self.unet = unet
+        self.controlnet = diffusers.ControlNetModel.from_unet(unet.unet)
+        DefaultAttnProc = AttnProcessor2_0
+        if is_xformers_available():
+            DefaultAttnProc = XFormersAttnProcessor
+        self.controlnet.set_attn_processor(DefaultAttnProc())
+
+    def __getattr__(self, name: str):
+        try:
+            return super().__getattr__(name)
+        except AttributeError:
+            return getattr(self.unet, name)
+
+    def forward(self, sample, timestep, encoder_hidden_states, class_labels=None, *args, cross_attention_kwargs: dict, **kwargs):
+        cross_attention_kwargs = dict(cross_attention_kwargs)
+        control_depth = cross_attention_kwargs.pop('control_depth')
+        down_block_res_samples, mid_block_res_sample = self.controlnet(
+            sample,
+            timestep,
+            encoder_hidden_states=encoder_hidden_states,
+            controlnet_cond=control_depth,
+            return_dict=False,
+        )
+        return self.unet(
+            sample,
+            timestep,
+            encoder_hidden_states=encoder_hidden_states,
+            down_block_res_samples=down_block_res_samples,
+            mid_block_res_sample=mid_block_res_sample,
+            cross_attention_kwargs=cross_attention_kwargs
+        )
+
+
+class ModuleListDict(torch.nn.Module):
+    def __init__(self, procs: dict) -> None:
+        super().__init__()
+        self.keys = sorted(procs.keys())
+        self.values = torch.nn.ModuleList(procs[k] for k in self.keys)
+
+    def __getitem__(self, key):
+        return self.values[self.keys.index(key)]
+
+
+class SuperNet(torch.nn.Module):
+    def __init__(self, state_dict: Dict[str, torch.Tensor]):
+        super().__init__()
+        state_dict = OrderedDict((k, state_dict[k]) for k in sorted(state_dict.keys()))
+        self.layers = torch.nn.ModuleList(state_dict.values())
+        self.mapping = dict(enumerate(state_dict.keys()))
+        self.rev_mapping = {v: k for k, v in enumerate(state_dict.keys())}
+
+        # .processor for unet, .self_attn for text encoder
+        self.split_keys = [".processor", ".self_attn"]
+
+        # we add a hook to state_dict() and load_state_dict() so that the
+        # naming fits with `unet.attn_processors`
+        def map_to(module, state_dict, *args, **kwargs):
+            new_state_dict = {}
+            for key, value in state_dict.items():
+                num = int(key.split(".")[1])  # 0 is always "layers"
+                new_key = key.replace(f"layers.{num}", module.mapping[num])
+                new_state_dict[new_key] = value
+
+            return new_state_dict
+
+        def remap_key(key, state_dict):
+            for k in self.split_keys:
+                if k in key:
+                    return key.split(k)[0] + k
+            return key.split('.')[0]
+
+        def map_from(module, state_dict, *args, **kwargs):
+            all_keys = list(state_dict.keys())
+            for key in all_keys:
+                replace_key = remap_key(key, state_dict)
+                new_key = key.replace(replace_key, f"layers.{module.rev_mapping[replace_key]}")
+                state_dict[new_key] = state_dict[key]
+                del state_dict[key]
+
+        self._register_state_dict_hook(map_to)
+        self._register_load_state_dict_pre_hook(map_from, with_module=True)
+
+
+class Zero123PlusPipeline(diffusers.StableDiffusionPipeline):
+    tokenizer: transformers.CLIPTokenizer
+    text_encoder: transformers.CLIPTextModel
+    vision_encoder: transformers.CLIPVisionModelWithProjection
+
+    feature_extractor_clip: transformers.CLIPImageProcessor
+    unet: UNet2DConditionModel
+    scheduler: diffusers.schedulers.KarrasDiffusionSchedulers
+
+    vae: AutoencoderKL
+    ramping: nn.Linear
+
+    feature_extractor_vae: transformers.CLIPImageProcessor
+
+    depth_transforms_multi = transforms.Compose([
+        transforms.ToTensor(),
+        transforms.Normalize([0.5], [0.5])
+    ])
+
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        tokenizer: CLIPTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        vision_encoder: transformers.CLIPVisionModelWithProjection,
+        feature_extractor_clip: CLIPImageProcessor, 
+        feature_extractor_vae: CLIPImageProcessor,
+        ramping_coefficients: Optional[list] = None,
+        safety_checker=None,
+    ):
+        DiffusionPipeline.__init__(self)
+
+        self.register_modules(
+            vae=vae, text_encoder=text_encoder, tokenizer=tokenizer,
+            unet=unet, scheduler=scheduler, safety_checker=None,
+            vision_encoder=vision_encoder,
+            feature_extractor_clip=feature_extractor_clip,
+            feature_extractor_vae=feature_extractor_vae
+        )
+        self.register_to_config(ramping_coefficients=ramping_coefficients)
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
+        self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
+
+    def prepare(self):
+        train_sched = DDPMScheduler.from_config(self.scheduler.config)
+        if isinstance(self.unet, UNet2DConditionModel):
+            self.unet = RefOnlyNoisedUNet(self.unet, train_sched, self.scheduler).eval()
+
+    def add_controlnet(self):
+        self.unet = DepthControlUNet(self.unet)
+        return SuperNet(OrderedDict([('controlnet', self.unet.controlnet)]))
+
+    def encode_condition_image(self, image: torch.Tensor):
+        image = self.vae.encode(image).latent_dist.sample()
+        return image
+
+    @torch.no_grad()
+    def __call__(
+        self,
+        image: Image.Image = None,
+        prompt = "",
+        *args,
+        num_images_per_prompt: Optional[int] = 1,
+        guidance_scale=4.0,
+        depth_image: Image.Image = None,
+        output_type: Optional[str] = "pil",
+        width=640,
+        height=960,
+        num_inference_steps=28,
+        return_dict=True,
+        **kwargs
+    ):
+        self.prepare()
+        if image is None:
+            raise ValueError("Inputting embeddings not supported for this pipeline. Please pass an image.")
+        assert not isinstance(image, torch.Tensor)
+        image_1 = self.feature_extractor_vae(images=image, return_tensors="pt").pixel_values
+        image_2 = self.feature_extractor_clip(images=image, return_tensors="pt").pixel_values
+        if depth_image is not None and hasattr(self.unet, "controlnet"):
+            depth_image = self.depth_transforms_multi(depth_image).to(
+                device=self.unet.controlnet.device, dtype=self.unet.controlnet.dtype
+            )
+        image = image_1.to(device=self.vae.device, dtype=self.vae.dtype)
+        image_2 = image_2.to(device=self.vae.device, dtype=self.vae.dtype)
+        cond_lat = self.encode_condition_image(image)
+        if guidance_scale > 1:
+            negative_lat = self.encode_condition_image(torch.zeros_like(image))
+            cond_lat = torch.cat([negative_lat, cond_lat])
+        encoded = self.vision_encoder(image_2, output_hidden_states=False)
+        global_embeds = encoded.image_embeds
+        global_embeds = global_embeds.unsqueeze(-2)
+        
+        encoder_hidden_states = self._encode_prompt(
+            prompt,
+            self.device,
+            num_images_per_prompt,
+            False
+        )
+        ramp = global_embeds.new_tensor(self.config.ramping_coefficients).unsqueeze(-1)
+        encoder_hidden_states = encoder_hidden_states + global_embeds * ramp
+        cak = dict(cond_lat=cond_lat)
+        if hasattr(self.unet, "controlnet"):
+            cak['control_depth'] = depth_image
+        latents: torch.Tensor = super().__call__(
+            None,
+            *args,
+            cross_attention_kwargs=cak,
+            guidance_scale=guidance_scale,
+            num_images_per_prompt=num_images_per_prompt,
+            prompt_embeds=encoder_hidden_states,
+            num_inference_steps=num_inference_steps,
+            output_type='latent',
+            width=width,
+            height=height,
+            **kwargs
+        ).images
+        latents = unscale_latents(latents)
+        if not output_type == "latent":
+            image = unscale_image(self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0])
+        else:
+            image = latents
+
+        image = self.image_processor.postprocess(image, output_type=output_type)
+        if not return_dict:
+            return (image,)
+
+        return ImagePipelineOutput(images=image)

model_index.json

@@ -0,0 +1,119 @@
+{
+  "_class_name": "Zero123PlusPipeline",
+  "_diffusers_version": "0.17.1",
+  "feature_extractor_clip": [
+    "transformers",
+    "CLIPImageProcessor"
+  ],
+  "feature_extractor_vae": [
+    "transformers",
+    "CLIPImageProcessor"
+  ],
+  "ramping_coefficients": [
+    0.0,
+    0.2060057818889618,
+    0.18684479594230652,
+    0.24342191219329834,
+    0.18507817387580872,
+    0.1703828126192093,
+    0.15628913044929504,
+    0.14174538850784302,
+    0.13617539405822754,
+    0.13569170236587524,
+    0.1269884556531906,
+    0.1200924888253212,
+    0.12816639244556427,
+    0.13058121502399445,
+    0.14201879501342773,
+    0.15004529058933258,
+    0.1620427817106247,
+    0.17207716405391693,
+    0.18534132838249207,
+    0.20002241432666779,
+    0.21657466888427734,
+    0.22996725142002106,
+    0.24613411724567413,
+    0.25141021609306335,
+    0.26613450050354004,
+    0.271847128868103,
+    0.2850190997123718,
+    0.285749226808548,
+    0.2813953757286072,
+    0.29509517550468445,
+    0.30109965801239014,
+    0.31370124220848083,
+    0.3134534955024719,
+    0.3108579218387604,
+    0.32147032022476196,
+    0.33548328280448914,
+    0.3301997184753418,
+    0.3254660964012146,
+    0.3514464199542999,
+    0.35993096232414246,
+    0.3510829508304596,
+    0.37661612033843994,
+    0.3913513123989105,
+    0.42122599482536316,
+    0.3954688012599945,
+    0.4260983467102051,
+    0.479139506816864,
+    0.4588979482650757,
+    0.4873477816581726,
+    0.5095643401145935,
+    0.5133851170539856,
+    0.520708441734314,
+    0.5363377928733826,
+    0.5661528706550598,
+    0.5859065651893616,
+    0.6207258701324463,
+    0.6560986638069153,
+    0.6379964351654053,
+    0.6777164340019226,
+    0.6589891910552979,
+    0.7574057579040527,
+    0.7446827292442322,
+    0.7695522308349609,
+    0.8163619041442871,
+    0.9502472281455994,
+    0.9918442368507385,
+    0.9398387670516968,
+    1.005432367324829,
+    0.9295969605445862,
+    0.9899859428405762,
+    1.044832706451416,
+    1.0427014827728271,
+    1.0829696655273438,
+    1.0062562227249146,
+    1.0966323614120483,
+    1.0550328493118286,
+    1.2108079195022583
+  ],
+  "safety_checker": [
+    null,
+    null
+  ],
+  "scheduler": [
+    "diffusers",
+    "EulerAncestralDiscreteScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "CLIPTextModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DConditionModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "vision_encoder": [
+    "transformers",
+    "CLIPVisionModelWithProjection"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,14 @@
+{
+  "_class_name": "EulerAncestralDiscreteScheduler",
+  "_diffusers_version": "0.17.1",
+  "beta_end": 0.012,
+  "beta_schedule": "linear",
+  "beta_start": 0.00085,
+  "clip_sample": false,
+  "num_train_timesteps": 1000,
+  "prediction_type": "v_prediction",
+  "set_alpha_to_one": false,
+  "skip_prk_steps": true,
+  "steps_offset": 1,
+  "trained_betas": null
+}

text_encoder/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "D:\\.cache\\huggingface\\hub\\models--stabilityai--stable-diffusion-2\\snapshots\\1e128c8891e52218b74cde8f26dbfc701cb99d79\\text_encoder",
+  "architectures": [
+    "CLIPTextModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_size": 1024,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 23,
+  "pad_token_id": 1,
+  "projection_dim": 512,
+  "torch_dtype": "float16",
+  "transformers_version": "4.29.0",
+  "vocab_size": 49408
+}

text_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2a06cf32cf585d03b55fef302142a5321b761ec440113925f64f4ceaffc7730
+size 680904225

tokenizer/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "!",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,33 @@
+{
+  "add_prefix_space": false,
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": true,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "errors": "replace",
+  "model_max_length": 77,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

unet/config.json

@@ -0,0 +1,67 @@
+{
+  "_class_name": "UNet2DConditionModel",
+  "_diffusers_version": "0.17.1",
+  "_name_or_path": "D:\\.cache\\huggingface\\hub\\models--stabilityai--stable-diffusion-2\\snapshots\\1e128c8891e52218b74cde8f26dbfc701cb99d79\\unet",
+  "act_fn": "silu",
+  "addition_embed_type": null,
+  "addition_embed_type_num_heads": 64,
+  "attention_head_dim": [
+    5,
+    10,
+    20,
+    20
+  ],
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "class_embeddings_concat": false,
+  "conv_in_kernel": 3,
+  "conv_out_kernel": 3,
+  "cross_attention_dim": 1024,
+  "cross_attention_norm": null,
+  "down_block_types": [
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "dual_cross_attention": false,
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 4,
+  "layers_per_block": 2,
+  "mid_block_only_cross_attention": null,
+  "mid_block_scale_factor": 1,
+  "mid_block_type": "UNetMidBlock2DCrossAttn",
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "only_cross_attention": false,
+  "out_channels": 4,
+  "projection_class_embeddings_input_dim": null,
+  "resnet_out_scale_factor": 1.0,
+  "resnet_skip_time_act": false,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 96,
+  "time_cond_proj_dim": null,
+  "time_embedding_act_fn": null,
+  "time_embedding_dim": null,
+  "time_embedding_type": "positional",
+  "timestep_post_act": null,
+  "up_block_types": [
+    "UpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
+  ],
+  "upcast_attention": false,
+  "use_linear_projection": true
+}

unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d78d78421d1feb6871d13e13e86ed8099628648d7d9c51ffca9015b7d5fa3c4
+size 1732056502

vae/config.json

@@ -0,0 +1,31 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.17.1",
+  "_name_or_path": "stabilityai/sd-vae-ft-mse",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 3,
+  "latent_channels": 4,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

vae/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cfdd672df17db3283633acb3721afc7735927293c2d3bd2bf64939a6dcd950e
+size 167407857

vision_encoder/config.json

@@ -0,0 +1,23 @@
+{
+  "_name_or_path": "stabilityai/stable-diffusion-2-1-unclip",
+  "architectures": [
+    "CLIPVisionModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_size": 1280,
+  "image_size": 224,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "model_type": "clip_vision_model",
+  "num_attention_heads": 16,
+  "num_channels": 3,
+  "num_hidden_layers": 32,
+  "patch_size": 14,
+  "projection_dim": 1024,
+  "torch_dtype": "float16",
+  "transformers_version": "4.29.0"
+}

vision_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c626d61a7660d2f86a1f0b5f74f513f93789a99469f1af641cc1f77810427f7
+size 1264335601