diff --git a/invokeai/app/invocations/flux_text_encoder.py b/invokeai/app/invocations/flux_text_encoder.py index 8e33a3f0cd2..5c0d0ef2ac7 100644 --- a/invokeai/app/invocations/flux_text_encoder.py +++ b/invokeai/app/invocations/flux_text_encoder.py @@ -1,6 +1,9 @@ import torch + + +from einops import repeat from diffusers.pipelines.flux.pipeline_flux import FluxPipeline -from transformers import CLIPTextModel, CLIPTokenizer, T5EncoderModel, T5TokenizerFast +from transformers import CLIPTextModel, CLIPTokenizer, T5EncoderModel, T5Tokenizer from invokeai.app.invocations.baseinvocation import BaseInvocation, invocation from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField @@ -9,6 +12,7 @@ from invokeai.app.services.shared.invocation_context import InvocationContext from invokeai.backend.stable_diffusion.diffusion.conditioning_data import ConditioningFieldData, FLUXConditioningInfo from invokeai.backend.util.devices import TorchDevice +from invokeai.backend.flux.modules.conditioner import HFEncoder @invocation( @@ -69,26 +73,15 @@ def _encode_prompt(self, context: InvocationContext) -> tuple[torch.Tensor, torc assert isinstance(clip_text_encoder, CLIPTextModel) assert isinstance(t5_text_encoder, T5EncoderModel) assert isinstance(clip_tokenizer, CLIPTokenizer) - assert isinstance(t5_tokenizer, T5TokenizerFast) + assert isinstance(t5_tokenizer, T5Tokenizer) + + clip_encoder = HFEncoder(clip_text_encoder, clip_tokenizer, True, 77) + t5_encoder = HFEncoder(t5_text_encoder, t5_tokenizer, False, max_seq_len) - pipeline = FluxPipeline( - scheduler=None, - vae=None, - text_encoder=clip_text_encoder, - tokenizer=clip_tokenizer, - text_encoder_2=t5_text_encoder, - tokenizer_2=t5_tokenizer, - transformer=None, - ) + prompt = [self.positive_prompt] + prompt_embeds = t5_encoder(prompt) - # prompt_embeds: T5 embeddings - # pooled_prompt_embeds: CLIP embeddings - prompt_embeds, pooled_prompt_embeds, _ = pipeline.encode_prompt( - prompt=self.positive_prompt, - prompt_2=self.positive_prompt, - device=TorchDevice.choose_torch_device(), - max_sequence_length=max_seq_len, - ) + pooled_prompt_embeds = clip_encoder(prompt) assert isinstance(prompt_embeds, torch.Tensor) assert isinstance(pooled_prompt_embeds, torch.Tensor) diff --git a/invokeai/app/invocations/flux_text_to_image.py b/invokeai/app/invocations/flux_text_to_image.py index bfb1484ed12..d976535d59b 100644 --- a/invokeai/app/invocations/flux_text_to_image.py +++ b/invokeai/app/invocations/flux_text_to_image.py @@ -85,7 +85,6 @@ def _run_diffusion( clip_embeddings: torch.Tensor, t5_embeddings: torch.Tensor, ): - scheduler_info = context.models.load(self.transformer.scheduler) transformer_info = context.models.load(self.transformer.transformer) # HACK(ryand): Manually empty the cache. Currently we don't check the size of the model before loading it from @@ -93,9 +92,8 @@ def _run_diffusion( # if the cache is not empty. # context.models._services.model_manager.load.ram_cache.make_room(24 * 2**30) - with transformer_info as transformer, scheduler_info as scheduler: + with transformer_info as transformer: assert isinstance(transformer, FluxTransformer2DModel) - assert isinstance(scheduler, FlowMatchEulerDiscreteScheduler) flux_pipeline_with_transformer = FluxPipeline( scheduler=scheduler, diff --git a/invokeai/app/invocations/model.py b/invokeai/app/invocations/model.py index c3902c1cb14..3908bef4da9 100644 --- a/invokeai/app/invocations/model.py +++ b/invokeai/app/invocations/model.py @@ -1,5 +1,6 @@ import copy -from typing import List, Optional +from time import sleep +from typing import List, Optional, Literal, Dict from pydantic import BaseModel, Field @@ -13,7 +14,8 @@ from invokeai.app.invocations.fields import FieldDescriptions, Input, InputField, OutputField, UIType from invokeai.app.services.shared.invocation_context import InvocationContext from invokeai.app.shared.models import FreeUConfig -from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelType, SubModelType +from invokeai.app.services.model_records import ModelRecordChanges +from invokeai.backend.model_manager.config import AnyModelConfig, BaseModelType, ModelType, SubModelType, ModelFormat class ModelIdentifierField(BaseModel): @@ -62,7 +64,6 @@ class CLIPField(BaseModel): class TransformerField(BaseModel): transformer: ModelIdentifierField = Field(description="Info to load Transformer submodel") - scheduler: ModelIdentifierField = Field(description="Info to load scheduler submodel") class T5EncoderField(BaseModel): @@ -131,6 +132,30 @@ def invoke(self, context: InvocationContext) -> ModelIdentifierOutput: return ModelIdentifierOutput(model=self.model) +T5_ENCODER_OPTIONS = Literal["base", "16b_quantized", "8b_quantized"] +T5_ENCODER_MAP: Dict[str, Dict[str, str]] = { + "base": { + "text_encoder_repo": "black-forest-labs/FLUX.1-schnell::text_encoder_2", + "tokenizer_repo": "black-forest-labs/FLUX.1-schnell::tokenizer_2", + "text_encoder_name": "FLUX.1-schnell_text_encoder_2", + "tokenizer_name": "FLUX.1-schnell_tokenizer_2", + "format": ModelFormat.T5Encoder, + }, + "8b_quantized": { + "text_encoder_repo": "hf_repo1", + "tokenizer_repo": "hf_repo1", + "text_encoder_name": "hf_repo1", + "tokenizer_name": "hf_repo1", + "format": ModelFormat.T5Encoder8b, + }, + "4b_quantized": { + "text_encoder_repo": "hf_repo2", + "tokenizer_repo": "hf_repo2", + "text_encoder_name": "hf_repo2", + "tokenizer_name": "hf_repo2", + "format": ModelFormat.T5Encoder8b, + }, +} @invocation_output("flux_model_loader_output") class FluxModelLoaderOutput(BaseInvocationOutput): @@ -151,29 +176,55 @@ class FluxModelLoaderInvocation(BaseInvocation): ui_type=UIType.FluxMainModel, input=Input.Direct, ) + + t5_encoder: T5_ENCODER_OPTIONS = InputField(description="The T5 Encoder model to use.") def invoke(self, context: InvocationContext) -> FluxModelLoaderOutput: model_key = self.model.key - # TODO: not found exceptions if not context.models.exists(model_key): raise Exception(f"Unknown model: {model_key}") - - transformer = self.model.model_copy(update={"submodel_type": SubModelType.Transformer}) - scheduler = self.model.model_copy(update={"submodel_type": SubModelType.Scheduler}) - tokenizer = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer}) - text_encoder = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder}) - tokenizer2 = self.model.model_copy(update={"submodel_type": SubModelType.Tokenizer2}) - text_encoder2 = self.model.model_copy(update={"submodel_type": SubModelType.TextEncoder2}) - vae = self.model.model_copy(update={"submodel_type": SubModelType.VAE}) + transformer = self._get_model(context, SubModelType.Transformer) + tokenizer = self._get_model(context, SubModelType.Tokenizer) + tokenizer2 = self._get_model(context, SubModelType.Tokenizer2) + clip_encoder = self._get_model(context, SubModelType.TextEncoder) + t5_encoder = self._get_model(context, SubModelType.TextEncoder2) + vae = self._install_model(context, SubModelType.VAE, "FLUX.1-schnell_ae", "black-forest-labs/FLUX.1-schnell::ae.safetensors", ModelFormat.Checkpoint, ModelType.VAE, BaseModelType.Flux) return FluxModelLoaderOutput( - transformer=TransformerField(transformer=transformer, scheduler=scheduler), - clip=CLIPField(tokenizer=tokenizer, text_encoder=text_encoder, loras=[], skipped_layers=0), - t5Encoder=T5EncoderField(tokenizer=tokenizer2, text_encoder=text_encoder2), + transformer=TransformerField(transformer=transformer), + clip=CLIPField(tokenizer=tokenizer, text_encoder=clip_encoder, loras=[], skipped_layers=0), + t5Encoder=T5EncoderField(tokenizer=tokenizer2, text_encoder=t5_encoder), vae=VAEField(vae=vae), ) + def _get_model(self, context: InvocationContext, submodel:SubModelType) -> ModelIdentifierField: + match(submodel): + case SubModelType.Transformer: + return self.model.model_copy(update={"submodel_type": SubModelType.Transformer}) + case submodel if submodel in [SubModelType.Tokenizer, SubModelType.TextEncoder]: + return self._install_model(context, submodel, "clip-vit-large-patch14", "openai/clip-vit-large-patch14", ModelFormat.Diffusers, ModelType.CLIPEmbed, BaseModelType.Any) + case SubModelType.TextEncoder2: + return self._install_model(context, submodel, T5_ENCODER_MAP[self.t5_encoder]["text_encoder_name"], T5_ENCODER_MAP[self.t5_encoder]["text_encoder_repo"], ModelFormat(T5_ENCODER_MAP[self.t5_encoder]["format"]), ModelType.T5Encoder, BaseModelType.Any) + case SubModelType.Tokenizer2: + return self._install_model(context, submodel, T5_ENCODER_MAP[self.t5_encoder]["tokenizer_name"], T5_ENCODER_MAP[self.t5_encoder]["tokenizer_repo"], ModelFormat(T5_ENCODER_MAP[self.t5_encoder]["format"]), ModelType.T5Encoder, BaseModelType.Any) + case _: + raise Exception(f"{submodel.value} is not a supported submodule for a flux model") + + def _install_model(self, context: InvocationContext, submodel:SubModelType, name: str, repo_id: str, format: ModelFormat, type: ModelType, base: BaseModelType): + if (models := context.models.search_by_attrs(name=name, base=base, type=type)): + if len(models) != 1: + raise Exception(f"Multiple models detected for selected model with name {name}") + return ModelIdentifierField.from_config(models[0]).model_copy(update={"submodel_type": submodel}) + else: + model_path = context.models.download_and_cache_model(repo_id) + config = ModelRecordChanges(name = name, base = base, type=type, format=format) + model_install_job = context.models.import_local_model(model_path=model_path, config=config) + while not model_install_job.in_terminal_state: + sleep(0.01) + if not model_install_job.config_out: + raise Exception(f"Failed to install {name}") + return ModelIdentifierField.from_config(model_install_job.config_out).model_copy(update={"submodel_type": submodel}) @invocation( "main_model_loader", diff --git a/invokeai/app/services/model_records/model_records_base.py b/invokeai/app/services/model_records/model_records_base.py index 46d11d4ddf2..9cc1486a019 100644 --- a/invokeai/app/services/model_records/model_records_base.py +++ b/invokeai/app/services/model_records/model_records_base.py @@ -77,6 +77,7 @@ class ModelRecordChanges(BaseModelExcludeNull): type: Optional[ModelType] = Field(description="Type of model", default=None) key: Optional[str] = Field(description="Database ID for this model", default=None) hash: Optional[str] = Field(description="hash of model file", default=None) + format: Optional[str] = Field(description="format of model file", default=None) trigger_phrases: Optional[set[str]] = Field(description="Set of trigger phrases for this model", default=None) default_settings: Optional[MainModelDefaultSettings | ControlAdapterDefaultSettings] = Field( description="Default settings for this model", default=None diff --git a/invokeai/app/services/model_records/model_records_sql.py b/invokeai/app/services/model_records/model_records_sql.py index 1d0780efe1f..d1ec0152429 100644 --- a/invokeai/app/services/model_records/model_records_sql.py +++ b/invokeai/app/services/model_records/model_records_sql.py @@ -301,7 +301,7 @@ def search_by_attr( for row in result: try: model_config = ModelConfigFactory.make_config(json.loads(row[0]), timestamp=row[1]) - except pydantic.ValidationError: + except pydantic.ValidationError as e: # We catch this error so that the app can still run if there are invalid model configs in the database. # One reason that an invalid model config might be in the database is if someone had to rollback from a # newer version of the app that added a new model type. diff --git a/invokeai/app/services/shared/invocation_context.py b/invokeai/app/services/shared/invocation_context.py index 01662335e46..9a5ac3fb5a9 100644 --- a/invokeai/app/services/shared/invocation_context.py +++ b/invokeai/app/services/shared/invocation_context.py @@ -13,6 +13,7 @@ from invokeai.app.services.image_records.image_records_common import ImageCategory, ResourceOrigin from invokeai.app.services.images.images_common import ImageDTO from invokeai.app.services.invocation_services import InvocationServices +from invokeai.app.services.model_records import ModelRecordChanges from invokeai.app.services.model_records.model_records_base import UnknownModelException from invokeai.app.util.step_callback import stable_diffusion_step_callback from invokeai.backend.model_manager.config import ( @@ -463,6 +464,20 @@ def download_and_cache_model( """ return self._services.model_manager.install.download_and_cache_model(source=source) + def import_local_model( + self, + model_path: Path, + config: Optional[ModelRecordChanges] = None, + access_token: Optional[str] = None, + inplace: Optional[bool] = False, + ): + """ + TODO: Fill out description of this method + """ + if not model_path.exists(): + raise Exception("Models provided to import_local_model must already exist on disk") + return self._services.model_manager.install.heuristic_import(str(model_path), config=config, access_token=access_token, inplace=inplace) + def load_local_model( self, model_path: Path, diff --git a/invokeai/backend/flux/math.py b/invokeai/backend/flux/math.py new file mode 100644 index 00000000000..71b91fa0f5a --- /dev/null +++ b/invokeai/backend/flux/math.py @@ -0,0 +1,30 @@ +import torch +from einops import rearrange +from torch import Tensor + + +def attention(q: Tensor, k: Tensor, v: Tensor, pe: Tensor) -> Tensor: + q, k = apply_rope(q, k, pe) + + x = torch.nn.functional.scaled_dot_product_attention(q, k, v) + x = rearrange(x, "B H L D -> B L (H D)") + + return x + + +def rope(pos: Tensor, dim: int, theta: int) -> Tensor: + assert dim % 2 == 0 + scale = torch.arange(0, dim, 2, dtype=torch.float64, device=pos.device) / dim + omega = 1.0 / (theta**scale) + out = torch.einsum("...n,d->...nd", pos, omega) + out = torch.stack([torch.cos(out), -torch.sin(out), torch.sin(out), torch.cos(out)], dim=-1) + out = rearrange(out, "b n d (i j) -> b n d i j", i=2, j=2) + return out.float() + + +def apply_rope(xq: Tensor, xk: Tensor, freqs_cis: Tensor) -> tuple[Tensor, Tensor]: + xq_ = xq.float().reshape(*xq.shape[:-1], -1, 1, 2) + xk_ = xk.float().reshape(*xk.shape[:-1], -1, 1, 2) + xq_out = freqs_cis[..., 0] * xq_[..., 0] + freqs_cis[..., 1] * xq_[..., 1] + xk_out = freqs_cis[..., 0] * xk_[..., 0] + freqs_cis[..., 1] * xk_[..., 1] + return xq_out.reshape(*xq.shape).type_as(xq), xk_out.reshape(*xk.shape).type_as(xk) \ No newline at end of file diff --git a/invokeai/backend/flux/model.py b/invokeai/backend/flux/model.py new file mode 100644 index 00000000000..2cb0aa102e7 --- /dev/null +++ b/invokeai/backend/flux/model.py @@ -0,0 +1,111 @@ +from dataclasses import dataclass + +import torch +from torch import Tensor, nn + +from invokeai.backend.flux.modules.layers import (DoubleStreamBlock, EmbedND, LastLayer, + MLPEmbedder, SingleStreamBlock, + timestep_embedding) + +@dataclass +class FluxParams: + in_channels: int + vec_in_dim: int + context_in_dim: int + hidden_size: int + mlp_ratio: float + num_heads: int + depth: int + depth_single_blocks: int + axes_dim: list[int] + theta: int + qkv_bias: bool + guidance_embed: bool + + +class Flux(nn.Module): + """ + Transformer model for flow matching on sequences. + """ + + def __init__(self, params: FluxParams): + super().__init__() + + self.params = params + self.in_channels = params.in_channels + self.out_channels = self.in_channels + if params.hidden_size % params.num_heads != 0: + raise ValueError( + f"Hidden size {params.hidden_size} must be divisible by num_heads {params.num_heads}" + ) + pe_dim = params.hidden_size // params.num_heads + if sum(params.axes_dim) != pe_dim: + raise ValueError(f"Got {params.axes_dim} but expected positional dim {pe_dim}") + self.hidden_size = params.hidden_size + self.num_heads = params.num_heads + self.pe_embedder = EmbedND(dim=pe_dim, theta=params.theta, axes_dim=params.axes_dim) + self.img_in = nn.Linear(self.in_channels, self.hidden_size, bias=True) + self.time_in = MLPEmbedder(in_dim=256, hidden_dim=self.hidden_size) + self.vector_in = MLPEmbedder(params.vec_in_dim, self.hidden_size) + self.guidance_in = ( + MLPEmbedder(in_dim=256, hidden_dim=self.hidden_size) if params.guidance_embed else nn.Identity() + ) + self.txt_in = nn.Linear(params.context_in_dim, self.hidden_size) + + self.double_blocks = nn.ModuleList( + [ + DoubleStreamBlock( + self.hidden_size, + self.num_heads, + mlp_ratio=params.mlp_ratio, + qkv_bias=params.qkv_bias, + ) + for _ in range(params.depth) + ] + ) + + self.single_blocks = nn.ModuleList( + [ + SingleStreamBlock(self.hidden_size, self.num_heads, mlp_ratio=params.mlp_ratio) + for _ in range(params.depth_single_blocks) + ] + ) + + self.final_layer = LastLayer(self.hidden_size, 1, self.out_channels) + + def forward( + self, + img: Tensor, + img_ids: Tensor, + txt: Tensor, + txt_ids: Tensor, + timesteps: Tensor, + y: Tensor, + guidance: Tensor | None = None, + ) -> Tensor: + if img.ndim != 3 or txt.ndim != 3: + raise ValueError("Input img and txt tensors must have 3 dimensions.") + + # running on sequences img + img = self.img_in(img) + vec = self.time_in(timestep_embedding(timesteps, 256)) + if self.params.guidance_embed: + if guidance is None: + raise ValueError("Didn't get guidance strength for guidance distilled model.") + vec = vec + self.guidance_in(timestep_embedding(guidance, 256)) + vec = vec + self.vector_in(y) + txt = self.txt_in(txt) + + ids = torch.cat((txt_ids, img_ids), dim=1) + pe = self.pe_embedder(ids) + + for block in self.double_blocks: + img, txt = block(img=img, txt=txt, vec=vec, pe=pe) + + img = torch.cat((txt, img), 1) + for block in self.single_blocks: + img = block(img, vec=vec, pe=pe) + img = img[:, txt.shape[1] :, ...] + + img = self.final_layer(img, vec) # (N, T, patch_size ** 2 * out_channels) + return img \ No newline at end of file diff --git a/invokeai/backend/flux/modules/autoencoder.py b/invokeai/backend/flux/modules/autoencoder.py new file mode 100644 index 00000000000..f6e072ecaaa --- /dev/null +++ b/invokeai/backend/flux/modules/autoencoder.py @@ -0,0 +1,312 @@ +from dataclasses import dataclass + +import torch +from einops import rearrange +from torch import Tensor, nn + + +@dataclass +class AutoEncoderParams: + resolution: int + in_channels: int + ch: int + out_ch: int + ch_mult: list[int] + num_res_blocks: int + z_channels: int + scale_factor: float + shift_factor: float + + +def swish(x: Tensor) -> Tensor: + return x * torch.sigmoid(x) + + +class AttnBlock(nn.Module): + def __init__(self, in_channels: int): + super().__init__() + self.in_channels = in_channels + + self.norm = nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True) + + self.q = nn.Conv2d(in_channels, in_channels, kernel_size=1) + self.k = nn.Conv2d(in_channels, in_channels, kernel_size=1) + self.v = nn.Conv2d(in_channels, in_channels, kernel_size=1) + self.proj_out = nn.Conv2d(in_channels, in_channels, kernel_size=1) + + def attention(self, h_: Tensor) -> Tensor: + h_ = self.norm(h_) + q = self.q(h_) + k = self.k(h_) + v = self.v(h_) + + b, c, h, w = q.shape + q = rearrange(q, "b c h w -> b 1 (h w) c").contiguous() + k = rearrange(k, "b c h w -> b 1 (h w) c").contiguous() + v = rearrange(v, "b c h w -> b 1 (h w) c").contiguous() + h_ = nn.functional.scaled_dot_product_attention(q, k, v) + + return rearrange(h_, "b 1 (h w) c -> b c h w", h=h, w=w, c=c, b=b) + + def forward(self, x: Tensor) -> Tensor: + return x + self.proj_out(self.attention(x)) + + +class ResnetBlock(nn.Module): + def __init__(self, in_channels: int, out_channels: int): + super().__init__() + self.in_channels = in_channels + out_channels = in_channels if out_channels is None else out_channels + self.out_channels = out_channels + + self.norm1 = nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True) + self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1) + self.norm2 = nn.GroupNorm(num_groups=32, num_channels=out_channels, eps=1e-6, affine=True) + self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1) + if self.in_channels != self.out_channels: + self.nin_shortcut = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0) + + def forward(self, x): + h = x + h = self.norm1(h) + h = swish(h) + h = self.conv1(h) + + h = self.norm2(h) + h = swish(h) + h = self.conv2(h) + + if self.in_channels != self.out_channels: + x = self.nin_shortcut(x) + + return x + h + + +class Downsample(nn.Module): + def __init__(self, in_channels: int): + super().__init__() + # no asymmetric padding in torch conv, must do it ourselves + self.conv = nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=2, padding=0) + + def forward(self, x: Tensor): + pad = (0, 1, 0, 1) + x = nn.functional.pad(x, pad, mode="constant", value=0) + x = self.conv(x) + return x + + +class Upsample(nn.Module): + def __init__(self, in_channels: int): + super().__init__() + self.conv = nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=1, padding=1) + + def forward(self, x: Tensor): + x = nn.functional.interpolate(x, scale_factor=2.0, mode="nearest") + x = self.conv(x) + return x + + +class Encoder(nn.Module): + def __init__( + self, + resolution: int, + in_channels: int, + ch: int, + ch_mult: list[int], + num_res_blocks: int, + z_channels: int, + ): + super().__init__() + self.ch = ch + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + self.resolution = resolution + self.in_channels = in_channels + # downsampling + self.conv_in = nn.Conv2d(in_channels, self.ch, kernel_size=3, stride=1, padding=1) + + curr_res = resolution + in_ch_mult = (1,) + tuple(ch_mult) + self.in_ch_mult = in_ch_mult + self.down = nn.ModuleList() + block_in = self.ch + for i_level in range(self.num_resolutions): + block = nn.ModuleList() + attn = nn.ModuleList() + block_in = ch * in_ch_mult[i_level] + block_out = ch * ch_mult[i_level] + for _ in range(self.num_res_blocks): + block.append(ResnetBlock(in_channels=block_in, out_channels=block_out)) + block_in = block_out + down = nn.Module() + down.block = block + down.attn = attn + if i_level != self.num_resolutions - 1: + down.downsample = Downsample(block_in) + curr_res = curr_res // 2 + self.down.append(down) + + # middle + self.mid = nn.Module() + self.mid.block_1 = ResnetBlock(in_channels=block_in, out_channels=block_in) + self.mid.attn_1 = AttnBlock(block_in) + self.mid.block_2 = ResnetBlock(in_channels=block_in, out_channels=block_in) + + # end + self.norm_out = nn.GroupNorm(num_groups=32, num_channels=block_in, eps=1e-6, affine=True) + self.conv_out = nn.Conv2d(block_in, 2 * z_channels, kernel_size=3, stride=1, padding=1) + + def forward(self, x: Tensor) -> Tensor: + # downsampling + hs = [self.conv_in(x)] + for i_level in range(self.num_resolutions): + for i_block in range(self.num_res_blocks): + h = self.down[i_level].block[i_block](hs[-1]) + if len(self.down[i_level].attn) > 0: + h = self.down[i_level].attn[i_block](h) + hs.append(h) + if i_level != self.num_resolutions - 1: + hs.append(self.down[i_level].downsample(hs[-1])) + + # middle + h = hs[-1] + h = self.mid.block_1(h) + h = self.mid.attn_1(h) + h = self.mid.block_2(h) + # end + h = self.norm_out(h) + h = swish(h) + h = self.conv_out(h) + return h + + +class Decoder(nn.Module): + def __init__( + self, + ch: int, + out_ch: int, + ch_mult: list[int], + num_res_blocks: int, + in_channels: int, + resolution: int, + z_channels: int, + ): + super().__init__() + self.ch = ch + self.num_resolutions = len(ch_mult) + self.num_res_blocks = num_res_blocks + self.resolution = resolution + self.in_channels = in_channels + self.ffactor = 2 ** (self.num_resolutions - 1) + + # compute in_ch_mult, block_in and curr_res at lowest res + block_in = ch * ch_mult[self.num_resolutions - 1] + curr_res = resolution // 2 ** (self.num_resolutions - 1) + self.z_shape = (1, z_channels, curr_res, curr_res) + + # z to block_in + self.conv_in = nn.Conv2d(z_channels, block_in, kernel_size=3, stride=1, padding=1) + + # middle + self.mid = nn.Module() + self.mid.block_1 = ResnetBlock(in_channels=block_in, out_channels=block_in) + self.mid.attn_1 = AttnBlock(block_in) + self.mid.block_2 = ResnetBlock(in_channels=block_in, out_channels=block_in) + + # upsampling + self.up = nn.ModuleList() + for i_level in reversed(range(self.num_resolutions)): + block = nn.ModuleList() + attn = nn.ModuleList() + block_out = ch * ch_mult[i_level] + for _ in range(self.num_res_blocks + 1): + block.append(ResnetBlock(in_channels=block_in, out_channels=block_out)) + block_in = block_out + up = nn.Module() + up.block = block + up.attn = attn + if i_level != 0: + up.upsample = Upsample(block_in) + curr_res = curr_res * 2 + self.up.insert(0, up) # prepend to get consistent order + + # end + self.norm_out = nn.GroupNorm(num_groups=32, num_channels=block_in, eps=1e-6, affine=True) + self.conv_out = nn.Conv2d(block_in, out_ch, kernel_size=3, stride=1, padding=1) + + def forward(self, z: Tensor) -> Tensor: + # z to block_in + h = self.conv_in(z) + + # middle + h = self.mid.block_1(h) + h = self.mid.attn_1(h) + h = self.mid.block_2(h) + + # upsampling + for i_level in reversed(range(self.num_resolutions)): + for i_block in range(self.num_res_blocks + 1): + h = self.up[i_level].block[i_block](h) + if len(self.up[i_level].attn) > 0: + h = self.up[i_level].attn[i_block](h) + if i_level != 0: + h = self.up[i_level].upsample(h) + + # end + h = self.norm_out(h) + h = swish(h) + h = self.conv_out(h) + return h + + +class DiagonalGaussian(nn.Module): + def __init__(self, sample: bool = True, chunk_dim: int = 1): + super().__init__() + self.sample = sample + self.chunk_dim = chunk_dim + + def forward(self, z: Tensor) -> Tensor: + mean, logvar = torch.chunk(z, 2, dim=self.chunk_dim) + if self.sample: + std = torch.exp(0.5 * logvar) + return mean + std * torch.randn_like(mean) + else: + return mean + + +class AutoEncoder(nn.Module): + def __init__(self, params: AutoEncoderParams): + super().__init__() + self.encoder = Encoder( + resolution=params.resolution, + in_channels=params.in_channels, + ch=params.ch, + ch_mult=params.ch_mult, + num_res_blocks=params.num_res_blocks, + z_channels=params.z_channels, + ) + self.decoder = Decoder( + resolution=params.resolution, + in_channels=params.in_channels, + ch=params.ch, + out_ch=params.out_ch, + ch_mult=params.ch_mult, + num_res_blocks=params.num_res_blocks, + z_channels=params.z_channels, + ) + self.reg = DiagonalGaussian() + + self.scale_factor = params.scale_factor + self.shift_factor = params.shift_factor + + def encode(self, x: Tensor) -> Tensor: + z = self.reg(self.encoder(x)) + z = self.scale_factor * (z - self.shift_factor) + return z + + def decode(self, z: Tensor) -> Tensor: + z = z / self.scale_factor + self.shift_factor + return self.decoder(z) + + def forward(self, x: Tensor) -> Tensor: + return self.decode(self.encode(x)) \ No newline at end of file diff --git a/invokeai/backend/flux/modules/conditioner.py b/invokeai/backend/flux/modules/conditioner.py new file mode 100644 index 00000000000..2a9e17c20e3 --- /dev/null +++ b/invokeai/backend/flux/modules/conditioner.py @@ -0,0 +1,30 @@ +from torch import Tensor, nn +from transformers import (PreTrainedModel, PreTrainedTokenizer) + +class HFEncoder(nn.Module): + def __init__(self, encoder: PreTrainedModel, tokenizer: PreTrainedTokenizer, is_clip: bool, max_length: int): + super().__init__() + self.max_length = max_length + self.is_clip = is_clip + self.output_key = "pooler_output" if self.is_clip else "last_hidden_state" + self.tokenizer = tokenizer + self.hf_module = encoder + self.hf_module = self.hf_module.eval().requires_grad_(False) + + def forward(self, text: list[str]) -> Tensor: + batch_encoding = self.tokenizer( + text, + truncation=True, + max_length=self.max_length, + return_length=False, + return_overflowing_tokens=False, + padding="max_length", + return_tensors="pt", + ) + + outputs = self.hf_module( + input_ids=batch_encoding["input_ids"].to(self.hf_module.device), + attention_mask=None, + output_hidden_states=False, + ) + return outputs[self.output_key] \ No newline at end of file diff --git a/invokeai/backend/flux/modules/layers.py b/invokeai/backend/flux/modules/layers.py new file mode 100644 index 00000000000..cb4eee0c2d7 --- /dev/null +++ b/invokeai/backend/flux/modules/layers.py @@ -0,0 +1,253 @@ +import math +from dataclasses import dataclass + +import torch +from einops import rearrange +from torch import Tensor, nn + +from ..math import attention, rope + + +class EmbedND(nn.Module): + def __init__(self, dim: int, theta: int, axes_dim: list[int]): + super().__init__() + self.dim = dim + self.theta = theta + self.axes_dim = axes_dim + + def forward(self, ids: Tensor) -> Tensor: + n_axes = ids.shape[-1] + emb = torch.cat( + [rope(ids[..., i], self.axes_dim[i], self.theta) for i in range(n_axes)], + dim=-3, + ) + + return emb.unsqueeze(1) + + +def timestep_embedding(t: Tensor, dim, max_period=10000, time_factor: float = 1000.0): + """ + Create sinusoidal timestep embeddings. + :param t: a 1-D Tensor of N indices, one per batch element. + These may be fractional. + :param dim: the dimension of the output. + :param max_period: controls the minimum frequency of the embeddings. + :return: an (N, D) Tensor of positional embeddings. + """ + t = time_factor * t + half = dim // 2 + freqs = torch.exp(-math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32) / half).to( + t.device + ) + + args = t[:, None].float() * freqs[None] + embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1) + if dim % 2: + embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1) + if torch.is_floating_point(t): + embedding = embedding.to(t) + return embedding + + +class MLPEmbedder(nn.Module): + def __init__(self, in_dim: int, hidden_dim: int): + super().__init__() + self.in_layer = nn.Linear(in_dim, hidden_dim, bias=True) + self.silu = nn.SiLU() + self.out_layer = nn.Linear(hidden_dim, hidden_dim, bias=True) + + def forward(self, x: Tensor) -> Tensor: + return self.out_layer(self.silu(self.in_layer(x))) + + +class RMSNorm(torch.nn.Module): + def __init__(self, dim: int): + super().__init__() + self.scale = nn.Parameter(torch.ones(dim)) + + def forward(self, x: Tensor): + x_dtype = x.dtype + x = x.float() + rrms = torch.rsqrt(torch.mean(x**2, dim=-1, keepdim=True) + 1e-6) + return (x * rrms).to(dtype=x_dtype) * self.scale + + +class QKNorm(torch.nn.Module): + def __init__(self, dim: int): + super().__init__() + self.query_norm = RMSNorm(dim) + self.key_norm = RMSNorm(dim) + + def forward(self, q: Tensor, k: Tensor, v: Tensor) -> tuple[Tensor, Tensor]: + q = self.query_norm(q) + k = self.key_norm(k) + return q.to(v), k.to(v) + + +class SelfAttention(nn.Module): + def __init__(self, dim: int, num_heads: int = 8, qkv_bias: bool = False): + super().__init__() + self.num_heads = num_heads + head_dim = dim // num_heads + + self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias) + self.norm = QKNorm(head_dim) + self.proj = nn.Linear(dim, dim) + + def forward(self, x: Tensor, pe: Tensor) -> Tensor: + qkv = self.qkv(x) + q, k, v = rearrange(qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + q, k = self.norm(q, k, v) + x = attention(q, k, v, pe=pe) + x = self.proj(x) + return x + + +@dataclass +class ModulationOut: + shift: Tensor + scale: Tensor + gate: Tensor + + +class Modulation(nn.Module): + def __init__(self, dim: int, double: bool): + super().__init__() + self.is_double = double + self.multiplier = 6 if double else 3 + self.lin = nn.Linear(dim, self.multiplier * dim, bias=True) + + def forward(self, vec: Tensor) -> tuple[ModulationOut, ModulationOut | None]: + out = self.lin(nn.functional.silu(vec))[:, None, :].chunk(self.multiplier, dim=-1) + + return ( + ModulationOut(*out[:3]), + ModulationOut(*out[3:]) if self.is_double else None, + ) + + +class DoubleStreamBlock(nn.Module): + def __init__(self, hidden_size: int, num_heads: int, mlp_ratio: float, qkv_bias: bool = False): + super().__init__() + + mlp_hidden_dim = int(hidden_size * mlp_ratio) + self.num_heads = num_heads + self.hidden_size = hidden_size + self.img_mod = Modulation(hidden_size, double=True) + self.img_norm1 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + self.img_attn = SelfAttention(dim=hidden_size, num_heads=num_heads, qkv_bias=qkv_bias) + + self.img_norm2 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + self.img_mlp = nn.Sequential( + nn.Linear(hidden_size, mlp_hidden_dim, bias=True), + nn.GELU(approximate="tanh"), + nn.Linear(mlp_hidden_dim, hidden_size, bias=True), + ) + + self.txt_mod = Modulation(hidden_size, double=True) + self.txt_norm1 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + self.txt_attn = SelfAttention(dim=hidden_size, num_heads=num_heads, qkv_bias=qkv_bias) + + self.txt_norm2 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + self.txt_mlp = nn.Sequential( + nn.Linear(hidden_size, mlp_hidden_dim, bias=True), + nn.GELU(approximate="tanh"), + nn.Linear(mlp_hidden_dim, hidden_size, bias=True), + ) + + def forward(self, img: Tensor, txt: Tensor, vec: Tensor, pe: Tensor) -> tuple[Tensor, Tensor]: + img_mod1, img_mod2 = self.img_mod(vec) + txt_mod1, txt_mod2 = self.txt_mod(vec) + + # prepare image for attention + img_modulated = self.img_norm1(img) + img_modulated = (1 + img_mod1.scale) * img_modulated + img_mod1.shift + img_qkv = self.img_attn.qkv(img_modulated) + img_q, img_k, img_v = rearrange(img_qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + img_q, img_k = self.img_attn.norm(img_q, img_k, img_v) + + # prepare txt for attention + txt_modulated = self.txt_norm1(txt) + txt_modulated = (1 + txt_mod1.scale) * txt_modulated + txt_mod1.shift + txt_qkv = self.txt_attn.qkv(txt_modulated) + txt_q, txt_k, txt_v = rearrange(txt_qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + txt_q, txt_k = self.txt_attn.norm(txt_q, txt_k, txt_v) + + # run actual attention + q = torch.cat((txt_q, img_q), dim=2) + k = torch.cat((txt_k, img_k), dim=2) + v = torch.cat((txt_v, img_v), dim=2) + + attn = attention(q, k, v, pe=pe) + txt_attn, img_attn = attn[:, : txt.shape[1]], attn[:, txt.shape[1] :] + + # calculate the img bloks + img = img + img_mod1.gate * self.img_attn.proj(img_attn) + img = img + img_mod2.gate * self.img_mlp((1 + img_mod2.scale) * self.img_norm2(img) + img_mod2.shift) + + # calculate the txt bloks + txt = txt + txt_mod1.gate * self.txt_attn.proj(txt_attn) + txt = txt + txt_mod2.gate * self.txt_mlp((1 + txt_mod2.scale) * self.txt_norm2(txt) + txt_mod2.shift) + return img, txt + + +class SingleStreamBlock(nn.Module): + """ + A DiT block with parallel linear layers as described in + https://arxiv.org/abs/2302.05442 and adapted modulation interface. + """ + + def __init__( + self, + hidden_size: int, + num_heads: int, + mlp_ratio: float = 4.0, + qk_scale: float | None = None, + ): + super().__init__() + self.hidden_dim = hidden_size + self.num_heads = num_heads + head_dim = hidden_size // num_heads + self.scale = qk_scale or head_dim**-0.5 + + self.mlp_hidden_dim = int(hidden_size * mlp_ratio) + # qkv and mlp_in + self.linear1 = nn.Linear(hidden_size, hidden_size * 3 + self.mlp_hidden_dim) + # proj and mlp_out + self.linear2 = nn.Linear(hidden_size + self.mlp_hidden_dim, hidden_size) + + self.norm = QKNorm(head_dim) + + self.hidden_size = hidden_size + self.pre_norm = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + + self.mlp_act = nn.GELU(approximate="tanh") + self.modulation = Modulation(hidden_size, double=False) + + def forward(self, x: Tensor, vec: Tensor, pe: Tensor) -> Tensor: + mod, _ = self.modulation(vec) + x_mod = (1 + mod.scale) * self.pre_norm(x) + mod.shift + qkv, mlp = torch.split(self.linear1(x_mod), [3 * self.hidden_size, self.mlp_hidden_dim], dim=-1) + + q, k, v = rearrange(qkv, "B L (K H D) -> K B H L D", K=3, H=self.num_heads) + q, k = self.norm(q, k, v) + + # compute attention + attn = attention(q, k, v, pe=pe) + # compute activation in mlp stream, cat again and run second linear layer + output = self.linear2(torch.cat((attn, self.mlp_act(mlp)), 2)) + return x + mod.gate * output + + +class LastLayer(nn.Module): + def __init__(self, hidden_size: int, patch_size: int, out_channels: int): + super().__init__() + self.norm_final = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6) + self.linear = nn.Linear(hidden_size, patch_size * patch_size * out_channels, bias=True) + self.adaLN_modulation = nn.Sequential(nn.SiLU(), nn.Linear(hidden_size, 2 * hidden_size, bias=True)) + + def forward(self, x: Tensor, vec: Tensor) -> Tensor: + shift, scale = self.adaLN_modulation(vec).chunk(2, dim=1) + x = (1 + scale[:, None, :]) * self.norm_final(x) + shift[:, None, :] + x = self.linear(x) + return x \ No newline at end of file diff --git a/invokeai/backend/model_manager/config.py b/invokeai/backend/model_manager/config.py index 29ef9536668..dfa6cef29b3 100644 --- a/invokeai/backend/model_manager/config.py +++ b/invokeai/backend/model_manager/config.py @@ -67,7 +67,9 @@ class ModelType(str, Enum): TextualInversion = "embedding" IPAdapter = "ip_adapter" CLIPVision = "clip_vision" + CLIPEmbed = "clip_embed" T2IAdapter = "t2i_adapter" + T5Encoder = "t5_encoder" SpandrelImageToImage = "spandrel_image_to_image" @@ -106,6 +108,9 @@ class ModelFormat(str, Enum): EmbeddingFile = "embedding_file" EmbeddingFolder = "embedding_folder" InvokeAI = "invokeai" + T5Encoder = "t5_encoder" + T5Encoder8b = "t5_encoder_8b" + T5Encoder4b = "t5_encoder_4b" class SchedulerPredictionType(str, Enum): @@ -207,6 +212,18 @@ class LoRAConfigBase(ModelConfigBase): trigger_phrases: Optional[set[str]] = Field(description="Set of trigger phrases for this model", default=None) +class T5EncoderConfigBase(ModelConfigBase): + type: Literal[ModelType.T5Encoder] = ModelType.T5Encoder + + +class T5EncoderConfig(T5EncoderConfigBase): + format: Literal[ModelFormat.T5Encoder] = ModelFormat.T5Encoder + + @staticmethod + def get_tag() -> Tag: + return Tag(f"{ModelType.T5Encoder.value}.{ModelFormat.T5Encoder.value}") + + class LoRALyCORISConfig(LoRAConfigBase): """Model config for LoRA/Lycoris models.""" @@ -352,6 +369,17 @@ def get_tag() -> Tag: return Tag(f"{ModelType.IPAdapter.value}.{ModelFormat.Checkpoint.value}") +class CLIPEmbedDiffusersConfig(DiffusersConfigBase): + """Model config for Clip Embeddings.""" + + type: Literal[ModelType.CLIPEmbed] = ModelType.CLIPEmbed + format: Literal[ModelFormat.Diffusers] = ModelFormat.Diffusers + + @staticmethod + def get_tag() -> Tag: + return Tag(f"{ModelType.CLIPEmbed.value}.{ModelFormat.Diffusers.value}") + + class CLIPVisionDiffusersConfig(DiffusersConfigBase): """Model config for CLIPVision.""" @@ -416,6 +444,7 @@ def get_model_discriminator_value(v: Any) -> str: Annotated[ControlNetCheckpointConfig, ControlNetCheckpointConfig.get_tag()], Annotated[LoRALyCORISConfig, LoRALyCORISConfig.get_tag()], Annotated[LoRADiffusersConfig, LoRADiffusersConfig.get_tag()], + Annotated[T5EncoderConfig, T5EncoderConfig.get_tag()], Annotated[TextualInversionFileConfig, TextualInversionFileConfig.get_tag()], Annotated[TextualInversionFolderConfig, TextualInversionFolderConfig.get_tag()], Annotated[IPAdapterInvokeAIConfig, IPAdapterInvokeAIConfig.get_tag()], @@ -423,6 +452,7 @@ def get_model_discriminator_value(v: Any) -> str: Annotated[T2IAdapterConfig, T2IAdapterConfig.get_tag()], Annotated[SpandrelImageToImageConfig, SpandrelImageToImageConfig.get_tag()], Annotated[CLIPVisionDiffusersConfig, CLIPVisionDiffusersConfig.get_tag()], + Annotated[CLIPEmbedDiffusersConfig, CLIPEmbedDiffusersConfig.get_tag()], ], Discriminator(get_model_discriminator_value), ] diff --git a/invokeai/backend/model_manager/load/model_loaders/flux.py b/invokeai/backend/model_manager/load/model_loaders/flux.py new file mode 100644 index 00000000000..7a028a55e10 --- /dev/null +++ b/invokeai/backend/model_manager/load/model_loaders/flux.py @@ -0,0 +1,159 @@ +# Copyright (c) 2024, Brandon W. Rising and the InvokeAI Development Team +"""Class for Flux model loading in InvokeAI.""" + +from pathlib import Path +import yaml + +from dataclasses import fields +from safetensors.torch import load_file +from typing import Optional, Any +from transformers import T5EncoderModel, T5Tokenizer + +from invokeai.backend.model_manager import ( + AnyModel, + AnyModelConfig, + BaseModelType, + ModelFormat, + ModelType, + SubModelType, +) +from invokeai.backend.model_manager.config import ( + CheckpointConfigBase, + MainCheckpointConfig, + CLIPEmbedDiffusersConfig, + T5EncoderConfig, + VAECheckpointConfig, +) +from invokeai.app.services.config.config_default import get_config +from invokeai.backend.model_manager.load.model_loader_registry import ModelLoaderRegistry +from invokeai.backend.model_manager.load.model_loaders.generic_diffusers import GenericDiffusersLoader +from invokeai.backend.util.silence_warnings import SilenceWarnings +from invokeai.backend.util.devices import TorchDevice +from invokeai.backend.flux.model import Flux, FluxParams +from invokeai.backend.flux.modules.autoencoder import AutoEncoderParams, AutoEncoder +from transformers import (CLIPTextModel, CLIPTokenizer, T5EncoderModel, + T5Tokenizer) + +app_config = get_config() + + +@ModelLoaderRegistry.register(base=BaseModelType.Flux, type=ModelType.VAE, format=ModelFormat.Checkpoint) +class FluxVAELoader(GenericDiffusersLoader): + """Class to load VAE models.""" + + def _load_model( + self, + config: AnyModelConfig, + submodel_type: Optional[SubModelType] = None, + ) -> AnyModel: + if isinstance(config, VAECheckpointConfig): + model_path = Path(config.path) + load_class = AutoEncoder + legacy_config_path = app_config.legacy_conf_path / config.config_path + config_path = legacy_config_path.as_posix() + with open(config_path, "r") as stream: + try: + flux_conf = yaml.safe_load(stream) + except: + raise + + dataclass_fields = {f.name for f in fields(AutoEncoderParams)} + filtered_data = {k: v for k, v in flux_conf['params']['ae_params'].items() if k in dataclass_fields} + params = AutoEncoderParams(**filtered_data) + + with SilenceWarnings(): + model = load_class(params).to(self._torch_dtype) + # load_sft doesn't support torch.device + sd = load_file(model_path, device=str(TorchDevice.choose_torch_device())) + model.load_state_dict(sd, strict=False, assign=True) + + return model + else: + return super()._load_model(config, submodel_type) + + +@ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.CLIPEmbed, format=ModelFormat.Diffusers) +class ClipCheckpointModel(GenericDiffusersLoader): + """Class to load main models.""" + + def _load_model( + self, + config: AnyModelConfig, + submodel_type: Optional[SubModelType] = None, + ) -> AnyModel: + if not isinstance(config, CLIPEmbedDiffusersConfig): + raise Exception("Only Checkpoint Flux models are currently supported.") + + match submodel_type: + case SubModelType.Tokenizer: + return CLIPTokenizer.from_pretrained(config.path, max_length=77) + case SubModelType.TextEncoder: + return CLIPTextModel.from_pretrained(config.path) + + raise Exception("Only Checkpoint Flux models are currently supported.") + +@ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.T5Encoder, format=ModelFormat.T5Encoder) +class T5EncoderCheckpointModel(GenericDiffusersLoader): + """Class to load main models.""" + + def _load_model( + self, + config: AnyModelConfig, + submodel_type: Optional[SubModelType] = None, + ) -> AnyModel: + if not isinstance(config, T5EncoderConfig): + raise Exception("Only Checkpoint Flux models are currently supported.") + + match submodel_type: + case SubModelType.Tokenizer2: + return T5Tokenizer.from_pretrained(Path(config.path), max_length=512) + case SubModelType.TextEncoder2: + return T5EncoderModel.from_pretrained(Path(config.path)) + + raise Exception("Only Checkpoint Flux models are currently supported.") + + +@ModelLoaderRegistry.register(base=BaseModelType.Flux, type=ModelType.Main, format=ModelFormat.Checkpoint) +class FluxCheckpointModel(GenericDiffusersLoader): + """Class to load main models.""" + + def _load_model( + self, + config: AnyModelConfig, + submodel_type: Optional[SubModelType] = None, + ) -> AnyModel: + if not isinstance(config, CheckpointConfigBase): + raise Exception("Only Checkpoint Flux models are currently supported.") + legacy_config_path = app_config.legacy_conf_path / config.config_path + config_path = legacy_config_path.as_posix() + with open(config_path, "r") as stream: + try: + flux_conf = yaml.safe_load(stream) + except: + raise + + match submodel_type: + case SubModelType.Transformer: + return self._load_from_singlefile(config, flux_conf) + + raise Exception("Only Checkpoint Flux models are currently supported.") + + def _load_from_singlefile( + self, + config: AnyModelConfig, + flux_conf: Any, + ) -> AnyModel: + assert isinstance(config, MainCheckpointConfig) + load_class = Flux + params = None + model_path = Path(config.path) + dataclass_fields = {f.name for f in fields(FluxParams)} + filtered_data = {k: v for k, v in flux_conf['params'].items() if k in dataclass_fields} + params = FluxParams(**filtered_data) + + with SilenceWarnings(): + model = load_class(params).to(self._torch_dtype) + # load_sft doesn't support torch.device + sd = load_file(model_path, device=str(TorchDevice.choose_torch_device())) + model.load_state_dict(sd, strict=False, assign=True) + return model diff --git a/invokeai/backend/model_manager/load/model_loaders/stable_diffusion.py b/invokeai/backend/model_manager/load/model_loaders/stable_diffusion.py index 33ce4abc4d4..e034e110115 100644 --- a/invokeai/backend/model_manager/load/model_loaders/stable_diffusion.py +++ b/invokeai/backend/model_manager/load/model_loaders/stable_diffusion.py @@ -36,8 +36,14 @@ } -@ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.Main, format=ModelFormat.Diffusers) -@ModelLoaderRegistry.register(base=BaseModelType.Any, type=ModelType.Main, format=ModelFormat.Checkpoint) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusion1, type=ModelType.Main, format=ModelFormat.Diffusers) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusion2, type=ModelType.Main, format=ModelFormat.Diffusers) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusionXL, type=ModelType.Main, format=ModelFormat.Diffusers) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusionXLRefiner, type=ModelType.Main, format=ModelFormat.Diffusers) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusion1, type=ModelType.Main, format=ModelFormat.Checkpoint) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusion2, type=ModelType.Main, format=ModelFormat.Checkpoint) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusionXL, type=ModelType.Main, format=ModelFormat.Checkpoint) +@ModelLoaderRegistry.register(base=BaseModelType.StableDiffusionXLRefiner, type=ModelType.Main, format=ModelFormat.Checkpoint) class StableDiffusionDiffusersModel(GenericDiffusersLoader): """Class to load main models.""" diff --git a/invokeai/backend/model_manager/load/model_util.py b/invokeai/backend/model_manager/load/model_util.py index f93b12a9b02..455f97d5484 100644 --- a/invokeai/backend/model_manager/load/model_util.py +++ b/invokeai/backend/model_manager/load/model_util.py @@ -9,7 +9,7 @@ import torch from diffusers.pipelines.pipeline_utils import DiffusionPipeline from diffusers.schedulers.scheduling_utils import SchedulerMixin -from transformers import CLIPTokenizer, T5TokenizerFast +from transformers import CLIPTokenizer, T5TokenizerFast, T5Tokenizer from invokeai.backend.image_util.grounding_dino.grounding_dino_pipeline import GroundingDinoPipeline from invokeai.backend.image_util.segment_anything.segment_anything_pipeline import SegmentAnythingPipeline @@ -50,7 +50,7 @@ def calc_model_size_by_data(logger: logging.Logger, model: AnyModel) -> int: return model.calc_size() elif isinstance( model, - (T5TokenizerFast,), + (T5TokenizerFast,T5Tokenizer,), ): return len(model) else: diff --git a/invokeai/backend/model_manager/probe.py b/invokeai/backend/model_manager/probe.py index 82053149ad1..a3a648806fc 100644 --- a/invokeai/backend/model_manager/probe.py +++ b/invokeai/backend/model_manager/probe.py @@ -56,7 +56,7 @@ }, BaseModelType.StableDiffusionXLRefiner: { ModelVariantType.Normal: "sd_xl_refiner.yaml", - }, + } } @@ -132,7 +132,7 @@ def probe( fields = {} model_path = model_path.resolve() - + format_type = ModelFormat.Diffusers if model_path.is_dir() else ModelFormat.Checkpoint model_info = None model_type = ModelType(fields["type"]) if "type" in fields and fields["type"] else None @@ -162,7 +162,7 @@ def probe( fields["description"] = ( fields.get("description") or f"{fields['base'].value} {model_type.value} model {fields['name']}" ) - fields["format"] = fields.get("format") or probe.get_format() + fields["format"] = ModelFormat(fields.get("format")) or probe.get_format() fields["hash"] = fields.get("hash") or ModelHash(algorithm=hash_algo).hash(model_path) fields["default_settings"] = fields.get("default_settings") @@ -223,7 +223,7 @@ def get_model_type_from_checkpoint(cls, model_path: Path, checkpoint: Optional[C ckpt = ckpt.get("state_dict", ckpt) for key in [str(k) for k in ckpt.keys()]: - if key.startswith(("cond_stage_model.", "first_stage_model.", "model.diffusion_model.")): + if key.startswith(("cond_stage_model.", "first_stage_model.", "model.diffusion_model.", "double_blocks.")): return ModelType.Main elif key.startswith(("encoder.conv_in", "decoder.conv_in")): return ModelType.VAE @@ -322,10 +322,13 @@ def _get_checkpoint_config_path( return possible_conf.absolute() if model_type is ModelType.Main: - config_file = LEGACY_CONFIGS[base_type][variant_type] - if isinstance(config_file, dict): # need another tier for sd-2.x models - config_file = config_file[prediction_type] - config_file = f"stable-diffusion/{config_file}" + if base_type == BaseModelType.Flux: + config_file="flux/flux1-schnell.yaml" + else: + config_file = LEGACY_CONFIGS[base_type][variant_type] + if isinstance(config_file, dict): # need another tier for sd-2.x models + config_file = config_file[prediction_type] + config_file = f"stable-diffusion/{config_file}" elif model_type is ModelType.ControlNet: config_file = ( "controlnet/cldm_v15.yaml" @@ -334,7 +337,9 @@ def _get_checkpoint_config_path( ) elif model_type is ModelType.VAE: config_file = ( - "stable-diffusion/v1-inference.yaml" + "flux/flux1-schnell.yaml" + if base_type is BaseModelType.Flux + else "stable-diffusion/v1-inference.yaml" if base_type is BaseModelType.StableDiffusion1 else "stable-diffusion/sd_xl_base.yaml" if base_type is BaseModelType.StableDiffusionXL @@ -421,7 +426,8 @@ def get_format(self) -> ModelFormat: def get_variant_type(self) -> ModelVariantType: model_type = ModelProbe.get_model_type_from_checkpoint(self.model_path, self.checkpoint) - if model_type != ModelType.Main: + base_type = self.get_base_type() + if model_type != ModelType.Main or base_type == BaseModelType.Flux: return ModelVariantType.Normal state_dict = self.checkpoint.get("state_dict") or self.checkpoint in_channels = state_dict["model.diffusion_model.input_blocks.0.0.weight"].shape[1] @@ -441,6 +447,8 @@ class PipelineCheckpointProbe(CheckpointProbeBase): def get_base_type(self) -> BaseModelType: checkpoint = self.checkpoint state_dict = self.checkpoint.get("state_dict") or checkpoint + if "double_blocks.0.img_attn.norm.key_norm.scale" in state_dict: + return BaseModelType.Flux key_name = "model.diffusion_model.input_blocks.2.1.transformer_blocks.0.attn2.to_k.weight" if key_name in state_dict and state_dict[key_name].shape[-1] == 768: return BaseModelType.StableDiffusion1 @@ -483,6 +491,7 @@ def get_base_type(self) -> BaseModelType: (r"xl", BaseModelType.StableDiffusionXL), (r"sd2", BaseModelType.StableDiffusion2), (r"vae", BaseModelType.StableDiffusion1), + (r"FLUX.1-schnell_ae", BaseModelType.Flux), ]: if re.search(regexp, self.model_path.name, re.IGNORECASE): return basetype @@ -627,10 +636,6 @@ def get_repo_variant(self) -> ModelRepoVariant: class PipelineFolderProbe(FolderProbeBase): def get_base_type(self) -> BaseModelType: - with open(f"{self.model_path}/model_index.json", "r") as file: - conf = json.load(file) - if "_class_name" in conf and conf.get("_class_name") == "FluxPipeline": - return BaseModelType.Flux with open(self.model_path / "unet" / "config.json", "r") as file: unet_conf = json.load(file) if unet_conf["cross_attention_dim"] == 768: @@ -718,6 +723,10 @@ def get_base_type(self) -> BaseModelType: return TextualInversionCheckpointProbe(path).get_base_type() +class T5EncoderFolderProbe(FolderProbeBase): + def get_format(self) -> ModelFormat: + return ModelFormat.T5Encoder + class ONNXFolderProbe(PipelineFolderProbe): def get_base_type(self) -> BaseModelType: # Due to the way the installer is set up, the configuration file for safetensors @@ -810,6 +819,11 @@ def get_base_type(self) -> BaseModelType: return BaseModelType.Any +class CLIPEmbedFolderProbe(FolderProbeBase): + def get_base_type(self) -> BaseModelType: + return BaseModelType.Any + + class SpandrelImageToImageFolderProbe(FolderProbeBase): def get_base_type(self) -> BaseModelType: raise NotImplementedError() @@ -840,8 +854,10 @@ def get_base_type(self) -> BaseModelType: ModelProbe.register_probe("diffusers", ModelType.VAE, VaeFolderProbe) ModelProbe.register_probe("diffusers", ModelType.LoRA, LoRAFolderProbe) ModelProbe.register_probe("diffusers", ModelType.TextualInversion, TextualInversionFolderProbe) +ModelProbe.register_probe("diffusers", ModelType.T5Encoder, T5EncoderFolderProbe) ModelProbe.register_probe("diffusers", ModelType.ControlNet, ControlNetFolderProbe) ModelProbe.register_probe("diffusers", ModelType.IPAdapter, IPAdapterFolderProbe) +ModelProbe.register_probe("diffusers", ModelType.CLIPEmbed, CLIPEmbedFolderProbe) ModelProbe.register_probe("diffusers", ModelType.CLIPVision, CLIPVisionFolderProbe) ModelProbe.register_probe("diffusers", ModelType.T2IAdapter, T2IAdapterFolderProbe) ModelProbe.register_probe("diffusers", ModelType.SpandrelImageToImage, SpandrelImageToImageFolderProbe) diff --git a/invokeai/configs/flux/flux1-dev.yaml b/invokeai/configs/flux/flux1-dev.yaml new file mode 100644 index 00000000000..3f76f11cd4e --- /dev/null +++ b/invokeai/configs/flux/flux1-dev.yaml @@ -0,0 +1,33 @@ +repo_id: "black-forest-labs/FLUX.1-dev" +repo_ae: "ae.safetensors" +max_length: 512 +params: + in_channels: 64 + vec_in_dim: 768 + context_in_dim: 4096 + hidden_size: 3072 + mlp_ratio: 4.0 + num_heads: 24 + depth: 19 + depth_single_blocks: 38 + axes_dim: + - 16 + - 56 + - 56 + theta: 10_000 + qkv_bias: True + guidance_embed: True + ae_params: + resolution: 256 + in_channels: 3 + ch: 128 + out_ch: 3 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + z_channels: 16 + scale_factor: 0.3611 + shift_factor: 0.1159 diff --git a/invokeai/configs/flux/flux1-schnell.yaml b/invokeai/configs/flux/flux1-schnell.yaml new file mode 100644 index 00000000000..bea1824e35a --- /dev/null +++ b/invokeai/configs/flux/flux1-schnell.yaml @@ -0,0 +1,34 @@ +repo_id: "black-forest-labs/FLUX.1-schnell" +repo_ae: "ae.safetensors" +t5_encoder: "google/t5-v1_1-xxl" +max_length: 512 +params: + in_channels: 64 + vec_in_dim: 768 + context_in_dim: 4096 + hidden_size: 3072 + mlp_ratio: 4.0 + num_heads: 24 + depth: 19 + depth_single_blocks: 38 + axes_dim: + - 16 + - 56 + - 56 + theta: 10_000 + qkv_bias: True + guidance_embed: False + ae_params: + resolution: 256 + in_channels: 3 + ch: 128 + out_ch: 3 + ch_mult: + - 1 + - 2 + - 4 + - 4 + num_res_blocks: 2 + z_channels: 16 + scale_factor: 0.3611 + shift_factor: 0.1159 diff --git a/invokeai/frontend/web/src/services/api/schema.ts b/invokeai/frontend/web/src/services/api/schema.ts index 74e1fb3a831..cfdca4354ec 100644 --- a/invokeai/frontend/web/src/services/api/schema.ts +++ b/invokeai/frontend/web/src/services/api/schema.ts @@ -4800,6 +4800,13 @@ export type components = { use_cache?: boolean; /** @description Flux model (Transformer, VAE, CLIP) to load */ model: components["schemas"]["ModelIdentifierField"]; + /** + * T5 Encoder + * @description The T5 Encoder model to use. + * @default null + * @enum {string} + */ + t5_encoder?: "base" | "16b_quantized" | "8b_quantized"; /** * type * @default flux_model_loader @@ -7675,154 +7682,154 @@ export type components = { project_id: string | null; }; InvocationOutputMap: { - mask_combine: components["schemas"]["ImageOutput"]; + lblend: components["schemas"]["LatentsOutput"]; + boolean_collection: components["schemas"]["BooleanCollectionOutput"]; + random_range: components["schemas"]["IntegerCollectionOutput"]; + merge_metadata: components["schemas"]["MetadataOutput"]; + string_split: components["schemas"]["String2Output"]; + string: components["schemas"]["StringOutput"]; + pair_tile_image: components["schemas"]["PairTileImageOutput"]; + tile_to_properties: components["schemas"]["TileToPropertiesOutput"]; + infill_lama: components["schemas"]["ImageOutput"]; + pidi_image_processor: components["schemas"]["ImageOutput"]; + lineart_image_processor: components["schemas"]["ImageOutput"]; + img_watermark: components["schemas"]["ImageOutput"]; + metadata_item: components["schemas"]["MetadataItemOutput"]; + integer: components["schemas"]["IntegerOutput"]; + img_conv: components["schemas"]["ImageOutput"]; + string_join: components["schemas"]["StringOutput"]; + lscale: components["schemas"]["LatentsOutput"]; + create_gradient_mask: components["schemas"]["GradientMaskOutput"]; + img_pad_crop: components["schemas"]["ImageOutput"]; + img_channel_offset: components["schemas"]["ImageOutput"]; + mask_edge: components["schemas"]["ImageOutput"]; + img_chan: components["schemas"]["ImageOutput"]; + string_join_three: components["schemas"]["StringOutput"]; + div: components["schemas"]["IntegerOutput"]; + merge_tiles_to_image: components["schemas"]["ImageOutput"]; + add: components["schemas"]["IntegerOutput"]; + leres_image_processor: components["schemas"]["ImageOutput"]; + img_lerp: components["schemas"]["ImageOutput"]; + t2i_adapter: components["schemas"]["T2IAdapterOutput"]; + float_to_int: components["schemas"]["IntegerOutput"]; + string_collection: components["schemas"]["StringCollectionOutput"]; + main_model_loader: components["schemas"]["ModelLoaderOutput"]; + heuristic_resize: components["schemas"]["ImageOutput"]; + img_channel_multiply: components["schemas"]["ImageOutput"]; + face_identifier: components["schemas"]["ImageOutput"]; img_ilerp: components["schemas"]["ImageOutput"]; - sdxl_model_loader: components["schemas"]["SDXLModelLoaderOutput"]; - dw_openpose_image_processor: components["schemas"]["ImageOutput"]; - infill_patchmatch: components["schemas"]["ImageOutput"]; - infill_tile: components["schemas"]["ImageOutput"]; - zoe_depth_image_processor: components["schemas"]["ImageOutput"]; - canny_image_processor: components["schemas"]["ImageOutput"]; - face_mask_detection: components["schemas"]["FaceMaskOutput"]; - core_metadata: components["schemas"]["MetadataOutput"]; - calculate_image_tiles_even_split: components["schemas"]["CalculateImageTilesOutput"]; - scheduler: components["schemas"]["SchedulerOutput"]; - clip_skip: components["schemas"]["CLIPSkipInvocationOutput"]; - img_nsfw: components["schemas"]["ImageOutput"]; - show_image: components["schemas"]["ImageOutput"]; - string_split_neg: components["schemas"]["StringPosNegOutput"]; - vae_loader: components["schemas"]["VAEOutput"]; - depth_anything_image_processor: components["schemas"]["ImageOutput"]; - canvas_paste_back: components["schemas"]["ImageOutput"]; - image_collection: components["schemas"]["ImageCollectionOutput"]; color: components["schemas"]["ColorOutput"]; - img_resize: components["schemas"]["ImageOutput"]; + rand_int: components["schemas"]["IntegerOutput"]; + infill_tile: components["schemas"]["ImageOutput"]; + sdxl_lora_loader: components["schemas"]["SDXLLoRALoaderOutput"]; ideal_size: components["schemas"]["IdealSizeOutput"]; - invert_tensor_mask: components["schemas"]["MaskOutput"]; - string_collection: components["schemas"]["StringCollectionOutput"]; - rectangle_mask: components["schemas"]["MaskOutput"]; - esrgan: components["schemas"]["ImageOutput"]; - image: components["schemas"]["ImageOutput"]; - sdxl_compel_prompt: components["schemas"]["ConditioningOutput"]; - metadata: components["schemas"]["MetadataOutput"]; - mlsd_image_processor: components["schemas"]["ImageOutput"]; + flux_model_loader: components["schemas"]["FluxModelLoaderOutput"]; + tile_image_processor: components["schemas"]["ImageOutput"]; color_correct: components["schemas"]["ImageOutput"]; - img_pad_crop: components["schemas"]["ImageOutput"]; - string_join_three: components["schemas"]["StringOutput"]; + lineart_anime_image_processor: components["schemas"]["ImageOutput"]; + latents_collection: components["schemas"]["LatentsCollectionOutput"]; + metadata: components["schemas"]["MetadataOutput"]; segment_anything: components["schemas"]["MaskOutput"]; + float: components["schemas"]["FloatOutput"]; + compel: components["schemas"]["ConditioningOutput"]; crop_latents: components["schemas"]["LatentsOutput"]; - integer: components["schemas"]["IntegerOutput"]; - img_channel_multiply: components["schemas"]["ImageOutput"]; - tiled_multi_diffusion_denoise_latents: components["schemas"]["LatentsOutput"]; - midas_depth_image_processor: components["schemas"]["ImageOutput"]; - img_scale: components["schemas"]["ImageOutput"]; - lora_collection_loader: components["schemas"]["LoRALoaderOutput"]; - cv_inpaint: components["schemas"]["ImageOutput"]; - sub: components["schemas"]["IntegerOutput"]; - flux_text_to_image: components["schemas"]["ImageOutput"]; - flux_model_loader: components["schemas"]["FluxModelLoaderOutput"]; - image_mask_to_tensor: components["schemas"]["MaskOutput"]; - pidi_image_processor: components["schemas"]["ImageOutput"]; - img_conv: components["schemas"]["ImageOutput"]; - unsharp_mask: components["schemas"]["ImageOutput"]; - t2i_adapter: components["schemas"]["T2IAdapterOutput"]; - rand_float: components["schemas"]["FloatOutput"]; - main_model_loader: components["schemas"]["ModelLoaderOutput"]; + tensor_mask_to_image: components["schemas"]["ImageOutput"]; + string_replace: components["schemas"]["StringOutput"]; + zoe_depth_image_processor: components["schemas"]["ImageOutput"]; + save_image: components["schemas"]["ImageOutput"]; + flux_text_encoder: components["schemas"]["ConditioningOutput"]; + sdxl_model_loader: components["schemas"]["SDXLModelLoaderOutput"]; + infill_rgba: components["schemas"]["ImageOutput"]; dynamic_prompt: components["schemas"]["StringCollectionOutput"]; - lresize: components["schemas"]["LatentsOutput"]; - float_collection: components["schemas"]["FloatCollectionOutput"]; - mediapipe_face_processor: components["schemas"]["ImageOutput"]; - lineart_image_processor: components["schemas"]["ImageOutput"]; - img_watermark: components["schemas"]["ImageOutput"]; - bounding_box: components["schemas"]["BoundingBoxOutput"]; - hed_image_processor: components["schemas"]["ImageOutput"]; - img_mul: components["schemas"]["ImageOutput"]; - img_lerp: components["schemas"]["ImageOutput"]; - compel: components["schemas"]["ConditioningOutput"]; - spandrel_image_to_image_autoscale: components["schemas"]["ImageOutput"]; - round_float: components["schemas"]["FloatOutput"]; - infill_lama: components["schemas"]["ImageOutput"]; + segment_anything_processor: components["schemas"]["ImageOutput"]; + lora_selector: components["schemas"]["LoRASelectorOutput"]; + content_shuffle_image_processor: components["schemas"]["ImageOutput"]; grounding_dino: components["schemas"]["BoundingBoxCollectionOutput"]; - model_identifier: components["schemas"]["ModelIdentifierOutput"]; - random_range: components["schemas"]["IntegerCollectionOutput"]; - float: components["schemas"]["FloatOutput"]; + cv_inpaint: components["schemas"]["ImageOutput"]; + mul: components["schemas"]["IntegerOutput"]; + float_range: components["schemas"]["FloatCollectionOutput"]; img_blur: components["schemas"]["ImageOutput"]; - img_chan: components["schemas"]["ImageOutput"]; - calculate_image_tiles_min_overlap: components["schemas"]["CalculateImageTilesOutput"]; - boolean: components["schemas"]["BooleanOutput"]; - metadata_item: components["schemas"]["MetadataItemOutput"]; - img_hue_adjust: components["schemas"]["ImageOutput"]; - string_split: components["schemas"]["String2Output"]; - float_to_int: components["schemas"]["IntegerOutput"]; - tensor_mask_to_image: components["schemas"]["ImageOutput"]; - seamless: components["schemas"]["SeamlessModeOutput"]; - create_denoise_mask: components["schemas"]["DenoiseMaskOutput"]; - string: components["schemas"]["StringOutput"]; - lblend: components["schemas"]["LatentsOutput"]; + core_metadata: components["schemas"]["MetadataOutput"]; + controlnet: components["schemas"]["ControlOutput"]; + image: components["schemas"]["ImageOutput"]; + infill_patchmatch: components["schemas"]["ImageOutput"]; + img_mul: components["schemas"]["ImageOutput"]; + prompt_from_file: components["schemas"]["StringCollectionOutput"]; + invert_tensor_mask: components["schemas"]["MaskOutput"]; + show_image: components["schemas"]["ImageOutput"]; + range: components["schemas"]["IntegerCollectionOutput"]; mask_from_id: components["schemas"]["ImageOutput"]; - latents_collection: components["schemas"]["LatentsCollectionOutput"]; - lineart_anime_image_processor: components["schemas"]["ImageOutput"]; - face_off: components["schemas"]["FaceOffOutput"]; - normalbae_image_processor: components["schemas"]["ImageOutput"]; - img_paste: components["schemas"]["ImageOutput"]; + range_of_size: components["schemas"]["IntegerCollectionOutput"]; + sdxl_compel_prompt: components["schemas"]["ConditioningOutput"]; + round_float: components["schemas"]["FloatOutput"]; + scheduler: components["schemas"]["SchedulerOutput"]; + image_collection: components["schemas"]["ImageCollectionOutput"]; + vae_loader: components["schemas"]["VAEOutput"]; infill_cv2: components["schemas"]["ImageOutput"]; - prompt_from_file: components["schemas"]["StringCollectionOutput"]; - segment_anything_processor: components["schemas"]["ImageOutput"]; - boolean_collection: components["schemas"]["BooleanCollectionOutput"]; + calculate_image_tiles_min_overlap: components["schemas"]["CalculateImageTilesOutput"]; + tiled_multi_diffusion_denoise_latents: components["schemas"]["LatentsOutput"]; + l2i: components["schemas"]["ImageOutput"]; + model_identifier: components["schemas"]["ModelIdentifierOutput"]; + mlsd_image_processor: components["schemas"]["ImageOutput"]; conditioning: components["schemas"]["ConditioningOutput"]; - flux_text_encoder: components["schemas"]["ConditioningOutput"]; - img_channel_offset: components["schemas"]["ImageOutput"]; - face_identifier: components["schemas"]["ImageOutput"]; + mask_combine: components["schemas"]["ImageOutput"]; + latents: components["schemas"]["LatentsOutput"]; + clip_skip: components["schemas"]["CLIPSkipInvocationOutput"]; + alpha_mask_to_tensor: components["schemas"]["MaskOutput"]; + tomask: components["schemas"]["ImageOutput"]; + create_denoise_mask: components["schemas"]["DenoiseMaskOutput"]; calculate_image_tiles: components["schemas"]["CalculateImageTilesOutput"]; - lora_loader: components["schemas"]["LoRALoaderOutput"]; - mul: components["schemas"]["IntegerOutput"]; - lscale: components["schemas"]["LatentsOutput"]; - integer_collection: components["schemas"]["IntegerCollectionOutput"]; - iterate: components["schemas"]["IterateInvocationOutput"]; - save_image: components["schemas"]["ImageOutput"]; - infill_rgba: components["schemas"]["ImageOutput"]; - content_shuffle_image_processor: components["schemas"]["ImageOutput"]; - tile_to_properties: components["schemas"]["TileToPropertiesOutput"]; + dw_openpose_image_processor: components["schemas"]["ImageOutput"]; + midas_depth_image_processor: components["schemas"]["ImageOutput"]; + boolean: components["schemas"]["BooleanOutput"]; + conditioning_collection: components["schemas"]["ConditioningCollectionOutput"]; + unsharp_mask: components["schemas"]["ImageOutput"]; + face_off: components["schemas"]["FaceOffOutput"]; + i2l: components["schemas"]["LatentsOutput"]; + normalbae_image_processor: components["schemas"]["ImageOutput"]; + img_hue_adjust: components["schemas"]["ImageOutput"]; + canny_image_processor: components["schemas"]["ImageOutput"]; + img_nsfw: components["schemas"]["ImageOutput"]; noise: components["schemas"]["NoiseOutput"]; - img_crop: components["schemas"]["ImageOutput"]; - freeu: components["schemas"]["UNetOutput"]; - integer_math: components["schemas"]["IntegerOutput"]; - div: components["schemas"]["IntegerOutput"]; - l2i: components["schemas"]["ImageOutput"]; - leres_image_processor: components["schemas"]["ImageOutput"]; - float_range: components["schemas"]["FloatCollectionOutput"]; - mask_edge: components["schemas"]["ImageOutput"]; - latents: components["schemas"]["LatentsOutput"]; - string_join: components["schemas"]["StringOutput"]; - add: components["schemas"]["IntegerOutput"]; + string_split_neg: components["schemas"]["StringPosNegOutput"]; sdxl_refiner_model_loader: components["schemas"]["SDXLRefinerModelLoaderOutput"]; - heuristic_resize: components["schemas"]["ImageOutput"]; - sdxl_lora_loader: components["schemas"]["SDXLLoRALoaderOutput"]; - color_map_image_processor: components["schemas"]["ImageOutput"]; - pair_tile_image: components["schemas"]["PairTileImageOutput"]; - i2l: components["schemas"]["LatentsOutput"]; - rand_int: components["schemas"]["IntegerOutput"]; - ip_adapter: components["schemas"]["IPAdapterOutput"]; - range_of_size: components["schemas"]["IntegerCollectionOutput"]; - merge_tiles_to_image: components["schemas"]["ImageOutput"]; + image_mask_to_tensor: components["schemas"]["MaskOutput"]; + lresize: components["schemas"]["LatentsOutput"]; + lora_loader: components["schemas"]["LoRALoaderOutput"]; + mediapipe_face_processor: components["schemas"]["ImageOutput"]; + esrgan: components["schemas"]["ImageOutput"]; sdxl_refiner_compel_prompt: components["schemas"]["ConditioningOutput"]; - controlnet: components["schemas"]["ControlOutput"]; - conditioning_collection: components["schemas"]["ConditioningCollectionOutput"]; - blank_image: components["schemas"]["ImageOutput"]; - sdxl_lora_collection_loader: components["schemas"]["SDXLLoRALoaderOutput"]; - create_gradient_mask: components["schemas"]["GradientMaskOutput"]; - collect: components["schemas"]["CollectInvocationOutput"]; - range: components["schemas"]["IntegerCollectionOutput"]; - merge_metadata: components["schemas"]["MetadataOutput"]; - string_replace: components["schemas"]["StringOutput"]; - tomask: components["schemas"]["ImageOutput"]; + integer_math: components["schemas"]["IntegerOutput"]; + seamless: components["schemas"]["SeamlessModeOutput"]; + img_paste: components["schemas"]["ImageOutput"]; + step_param_easing: components["schemas"]["FloatCollectionOutput"]; + hed_image_processor: components["schemas"]["ImageOutput"]; + img_scale: components["schemas"]["ImageOutput"]; + face_mask_detection: components["schemas"]["FaceMaskOutput"]; + img_resize: components["schemas"]["ImageOutput"]; denoise_latents: components["schemas"]["LatentsOutput"]; float_math: components["schemas"]["FloatOutput"]; + freeu: components["schemas"]["UNetOutput"]; + collect: components["schemas"]["CollectInvocationOutput"]; + blank_image: components["schemas"]["ImageOutput"]; + calculate_image_tiles_even_split: components["schemas"]["CalculateImageTilesOutput"]; + depth_anything_image_processor: components["schemas"]["ImageOutput"]; + bounding_box: components["schemas"]["BoundingBoxOutput"]; + color_map_image_processor: components["schemas"]["ImageOutput"]; spandrel_image_to_image: components["schemas"]["ImageOutput"]; - lora_selector: components["schemas"]["LoRASelectorOutput"]; - alpha_mask_to_tensor: components["schemas"]["MaskOutput"]; - tile_image_processor: components["schemas"]["ImageOutput"]; - step_param_easing: components["schemas"]["FloatCollectionOutput"]; + iterate: components["schemas"]["IterateInvocationOutput"]; + float_collection: components["schemas"]["FloatCollectionOutput"]; + sdxl_lora_collection_loader: components["schemas"]["SDXLLoRALoaderOutput"]; + rectangle_mask: components["schemas"]["MaskOutput"]; + flux_text_to_image: components["schemas"]["ImageOutput"]; + spandrel_image_to_image_autoscale: components["schemas"]["ImageOutput"]; + rand_float: components["schemas"]["FloatOutput"]; + integer_collection: components["schemas"]["IntegerCollectionOutput"]; + ip_adapter: components["schemas"]["IPAdapterOutput"]; + sub: components["schemas"]["IntegerOutput"]; + img_crop: components["schemas"]["ImageOutput"]; + lora_collection_loader: components["schemas"]["LoRALoaderOutput"]; + canvas_paste_back: components["schemas"]["ImageOutput"]; }; /** * InvocationStartedEvent @@ -9724,7 +9731,7 @@ export type components = { * @description Storage format of model. * @enum {string} */ - ModelFormat: "diffusers" | "checkpoint" | "lycoris" | "onnx" | "olive" | "embedding_file" | "embedding_folder" | "invokeai"; + ModelFormat: "diffusers" | "checkpoint" | "lycoris" | "onnx" | "olive" | "embedding_file" | "embedding_folder" | "invokeai" | "t5_encoder" | "t5_encoder_8b" | "t5_encoder_4b"; /** ModelIdentifierField */ ModelIdentifierField: { /** @@ -10024,7 +10031,7 @@ export type components = { * Config Out * @description After successful installation, this will hold the configuration object. */ - config_out?: (components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]) | null; + config_out?: (components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]) | null; /** * Inplace * @description Leave model in its current location; otherwise install under models directory @@ -10110,7 +10117,7 @@ export type components = { * Config * @description The model's config */ - config: components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + config: components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; /** * @description The submodel type, if any * @default null @@ -10131,7 +10138,7 @@ export type components = { * Config * @description The model's config */ - config: components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + config: components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; /** * @description The submodel type, if any * @default null @@ -10212,6 +10219,11 @@ export type components = { * @description hash of model file */ hash?: string | null; + /** + * Format + * @description format of model file + */ + format?: string | null; /** * Trigger Phrases * @description Set of trigger phrases for this model @@ -10254,7 +10266,7 @@ export type components = { * @description Model type. * @enum {string} */ - ModelType: "onnx" | "main" | "vae" | "lora" | "controlnet" | "embedding" | "ip_adapter" | "clip_vision" | "t2i_adapter" | "spandrel_image_to_image"; + ModelType: "onnx" | "main" | "vae" | "lora" | "controlnet" | "embedding" | "ip_adapter" | "clip_vision" | "t2i_adapter" | "t5_encoder" | "spandrel_image_to_image"; /** * ModelVariantType * @description Variant type. @@ -10267,7 +10279,7 @@ export type components = { */ ModelsList: { /** Models */ - models: (components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"])[]; + models: (components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"])[]; }; /** * Multiply Integers @@ -13338,6 +13350,67 @@ export type components = { */ type: "t2i_adapter_output"; }; + /** T5EncoderConfig */ + T5EncoderConfig: { + /** + * Key + * @description A unique key for this model. + */ + key: string; + /** + * Hash + * @description The hash of the model file(s). + */ + hash: string; + /** + * Path + * @description Path to the model on the filesystem. Relative paths are relative to the Invoke root directory. + */ + path: string; + /** + * Name + * @description Name of the model. + */ + name: string; + /** @description The base model. */ + base: components["schemas"]["BaseModelType"]; + /** + * Description + * @description Model description + */ + description?: string | null; + /** + * Source + * @description The original source of the model (path, URL or repo_id). + */ + source: string; + /** @description The type of source */ + source_type: components["schemas"]["ModelSourceType"]; + /** + * Source Api Response + * @description The original API response from the source, as stringified JSON. + */ + source_api_response?: string | null; + /** + * Cover Image + * @description Url for image to preview model + */ + cover_image?: string | null; + /** + * Type + * @default t5_encoder + * @constant + * @enum {string} + */ + type: "t5_encoder"; + /** + * Format + * @default t5_encoder + * @constant + * @enum {string} + */ + format: "t5_encoder"; + }; /** T5EncoderField */ T5EncoderField: { /** @description Info to load tokenizer submodel */ @@ -13780,8 +13853,6 @@ export type components = { TransformerField: { /** @description Info to load Transformer submodel */ transformer: components["schemas"]["ModelIdentifierField"]; - /** @description Info to load scheduler submodel */ - scheduler: components["schemas"]["ModelIdentifierField"]; }; /** * UIComponent @@ -14560,7 +14631,7 @@ export type operations = { /** @description Successful Response */ 200: { content: { - "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; }; }; /** @description Validation Error */ @@ -14586,7 +14657,7 @@ export type operations = { /** @description The model configuration was retrieved successfully */ 200: { content: { - "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; }; }; /** @description Bad request */ @@ -14668,7 +14739,7 @@ export type operations = { /** @description The model was updated successfully */ 200: { content: { - "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; }; }; /** @description Bad request */ @@ -15078,7 +15149,7 @@ export type operations = { /** @description Model converted successfully */ 200: { content: { - "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; + "application/json": components["schemas"]["MainDiffusersConfig"] | components["schemas"]["MainCheckpointConfig"] | components["schemas"]["VAEDiffusersConfig"] | components["schemas"]["VAECheckpointConfig"] | components["schemas"]["ControlNetDiffusersConfig"] | components["schemas"]["ControlNetCheckpointConfig"] | components["schemas"]["LoRALyCORISConfig"] | components["schemas"]["LoRADiffusersConfig"] | components["schemas"]["T5EncoderConfig"] | components["schemas"]["TextualInversionFileConfig"] | components["schemas"]["TextualInversionFolderConfig"] | components["schemas"]["IPAdapterInvokeAIConfig"] | components["schemas"]["IPAdapterCheckpointConfig"] | components["schemas"]["T2IAdapterConfig"] | components["schemas"]["SpandrelImageToImageConfig"] | components["schemas"]["CLIPVisionDiffusersConfig"]; }; }; /** @description Bad request */