Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-20 01:31:13 +02:00
Code Issues Packages Projects Releases Wiki Activity

modernize typing

Browse Source
This commit is contained in:
Vladimir Mandic
2026-02-19 09:15:37 +01:00
parent 7aded79e8a
commit bfe014f5da
222 changed files with 1538 additions and 1444 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/api/api.py
+16 -17
View File
@@ -1,4 +1,3 @@
from typing import List, Optional
from threading import Lock
from secrets import compare_digest
from fastapi import FastAPI, APIRouter, Depends, Request
@@ -19,7 +18,7 @@ class Api:
user, password = auth.split(":")
self.credentials[user.replace('"', '').strip()] = password.replace('"', '').strip()
if shared.cmd_opts.auth_file:
with open(shared.cmd_opts.auth_file, 'r', encoding="utf8") as file:
with open(shared.cmd_opts.auth_file, encoding="utf8") as file:
for line in file.readlines():
user, password = line.split(":")
self.credentials[user.replace('"', '').strip()] = password.replace('"', '').strip()
@@ -41,7 +40,7 @@ class Api:
self.add_api_route("/js", server.get_js, methods=["GET"], auth=False)
# server api
self.add_api_route("/sdapi/v1/motd", server.get_motd, methods=["GET"], response_model=str)
self.add_api_route("/sdapi/v1/log", server.get_log, methods=["GET"], response_model=List[str])
self.add_api_route("/sdapi/v1/log", server.get_log, methods=["GET"], response_model=list[str])
self.add_api_route("/sdapi/v1/log", server.post_log, methods=["POST"])
self.add_api_route("/sdapi/v1/start", self.get_session_start, methods=["GET"])
self.add_api_route("/sdapi/v1/version", server.get_version, methods=["GET"])
@@ -56,7 +55,7 @@ class Api:
self.add_api_route("/sdapi/v1/options", server.get_config, methods=["GET"], response_model=models.OptionsModel)
self.add_api_route("/sdapi/v1/options", server.set_config, methods=["POST"])
self.add_api_route("/sdapi/v1/cmd-flags", server.get_cmd_flags, methods=["GET"], response_model=models.FlagsModel)
self.add_api_route("/sdapi/v1/gpu", gpu.get_gpu_status, methods=["GET"], response_model=List[models.ResGPU])
self.add_api_route("/sdapi/v1/gpu", gpu.get_gpu_status, methods=["GET"], response_model=list[models.ResGPU])
# core api using locking
self.add_api_route("/sdapi/v1/txt2img", self.generate.post_text2img, methods=["POST"], response_model=models.ResTxt2Img)
@@ -71,21 +70,21 @@ class Api:
# api dealing with optional scripts
self.add_api_route("/sdapi/v1/scripts", script.get_scripts_list, methods=["GET"], response_model=models.ResScripts)
self.add_api_route("/sdapi/v1/script-info", script.get_script_info, methods=["GET"], response_model=List[models.ItemScript])
self.add_api_route("/sdapi/v1/script-info", script.get_script_info, methods=["GET"], response_model=list[models.ItemScript])
# enumerator api
self.add_api_route("/sdapi/v1/preprocessors", self.process.get_preprocess, methods=["GET"], response_model=List[process.ItemPreprocess])
self.add_api_route("/sdapi/v1/preprocessors", self.process.get_preprocess, methods=["GET"], response_model=list[process.ItemPreprocess])
self.add_api_route("/sdapi/v1/masking", self.process.get_mask, methods=["GET"], response_model=process.ItemMask)
self.add_api_route("/sdapi/v1/samplers", endpoints.get_samplers, methods=["GET"], response_model=List[models.ItemSampler])
self.add_api_route("/sdapi/v1/upscalers", endpoints.get_upscalers, methods=["GET"], response_model=List[models.ItemUpscaler])
self.add_api_route("/sdapi/v1/sd-models", endpoints.get_sd_models, methods=["GET"], response_model=List[models.ItemModel])
self.add_api_route("/sdapi/v1/controlnets", endpoints.get_controlnets, methods=["GET"], response_model=List[str])
self.add_api_route("/sdapi/v1/detailers", endpoints.get_detailers, methods=["GET"], response_model=List[models.ItemDetailer])
self.add_api_route("/sdapi/v1/prompt-styles", endpoints.get_prompt_styles, methods=["GET"], response_model=List[models.ItemStyle])
self.add_api_route("/sdapi/v1/samplers", endpoints.get_samplers, methods=["GET"], response_model=list[models.ItemSampler])
self.add_api_route("/sdapi/v1/upscalers", endpoints.get_upscalers, methods=["GET"], response_model=list[models.ItemUpscaler])
self.add_api_route("/sdapi/v1/sd-models", endpoints.get_sd_models, methods=["GET"], response_model=list[models.ItemModel])
self.add_api_route("/sdapi/v1/controlnets", endpoints.get_controlnets, methods=["GET"], response_model=list[str])
self.add_api_route("/sdapi/v1/detailers", endpoints.get_detailers, methods=["GET"], response_model=list[models.ItemDetailer])
self.add_api_route("/sdapi/v1/prompt-styles", endpoints.get_prompt_styles, methods=["GET"], response_model=list[models.ItemStyle])
self.add_api_route("/sdapi/v1/embeddings", endpoints.get_embeddings, methods=["GET"], response_model=models.ResEmbeddings)
self.add_api_route("/sdapi/v1/sd-vae", endpoints.get_sd_vaes, methods=["GET"], response_model=List[models.ItemVae])
self.add_api_route("/sdapi/v1/extensions", endpoints.get_extensions_list, methods=["GET"], response_model=List[models.ItemExtension])
self.add_api_route("/sdapi/v1/extra-networks", endpoints.get_extra_networks, methods=["GET"], response_model=List[models.ItemExtraNetwork])
self.add_api_route("/sdapi/v1/sd-vae", endpoints.get_sd_vaes, methods=["GET"], response_model=list[models.ItemVae])
self.add_api_route("/sdapi/v1/extensions", endpoints.get_extensions_list, methods=["GET"], response_model=list[models.ItemExtension])
self.add_api_route("/sdapi/v1/extra-networks", endpoints.get_extra_networks, methods=["GET"], response_model=list[models.ItemExtraNetwork])
# functional api
self.add_api_route("/sdapi/v1/png-info", endpoints.post_pnginfo, methods=["POST"], response_model=models.ResImageInfo)
@@ -96,7 +95,7 @@ class Api:
self.add_api_route("/sdapi/v1/reload-checkpoint", endpoints.post_reload_checkpoint, methods=["POST"])
self.add_api_route("/sdapi/v1/lock-checkpoint", endpoints.post_lock_checkpoint, methods=["POST"])
self.add_api_route("/sdapi/v1/refresh-vae", endpoints.post_refresh_vae, methods=["POST"])
self.add_api_route("/sdapi/v1/latents", endpoints.get_latent_history, methods=["GET"], response_model=List[str])
self.add_api_route("/sdapi/v1/latents", endpoints.get_latent_history, methods=["GET"], response_model=list[str])
self.add_api_route("/sdapi/v1/latents", endpoints.post_latent_history, methods=["POST"], response_model=int)
self.add_api_route("/sdapi/v1/modules", endpoints.get_modules, methods=["GET"])
self.add_api_route("/sdapi/v1/sampler", endpoints.get_sampler, methods=["GET"], response_model=dict)
@@ -146,7 +145,7 @@ class Api:
shared.log.error(f'API authentication: user="{credentials.username}"')
raise HTTPException(status_code=401, detail="Unauthorized", headers={"WWW-Authenticate": "Basic"})
def get_session_start(self, req: Request, agent: Optional[str] = None):
def get_session_start(self, req: Request, agent: str | None = None):
token = req.cookies.get("access-token") or req.cookies.get("access-token-unsecure")
user = self.app.tokens.get(token) if hasattr(self.app, 'tokens') else None
shared.log.info(f'Browser session: user={user} client={req.client.host} agent={agent}')
modules/api/caption.py
+67 -67
View File
@@ -25,7 +25,7 @@ Core processing logic is shared between direct and dispatch handlers via
``do_openclip``, ``do_tagger``, and ``do_vqa`` functions to avoid duplication.
"""
from typing import Optional, List, Union, Literal, Annotated
from typing import Literal, Annotated
from pydantic import BaseModel, Field # pylint: disable=no-name-in-module
from fastapi.exceptions import HTTPException
from modules import shared
@@ -49,21 +49,21 @@ class ReqCaption(BaseModel):
mode: str = Field(default="best", title="Mode", description="Caption mode. 'best': Most thorough analysis, slowest but highest quality. 'fast': Quick caption with minimal flavor terms. 'classic': Standard captioning with balanced quality and speed. 'caption': BLIP caption only, no CLIP flavor matching. 'negative': Generate terms suitable for use as a negative prompt.")
analyze: bool = Field(default=False, title="Analyze", description="If True, returns detailed image analysis breakdown (medium, artist, movement, trending, flavor) in addition to caption.")
# Advanced settings (optional per-request overrides)
max_length: Optional[int] = Field(default=None, title="Max Length", description="Maximum number of tokens in the generated caption.")
chunk_size: Optional[int] = Field(default=None, title="Chunk Size", description="Batch size for processing description candidates (flavors). Higher values speed up captioning but increase VRAM usage.")
min_flavors: Optional[int] = Field(default=None, title="Min Flavors", description="Minimum number of descriptive tags (flavors) to keep in the final prompt.")
max_flavors: Optional[int] = Field(default=None, title="Max Flavors", description="Maximum number of descriptive tags (flavors) to keep in the final prompt.")
flavor_count: Optional[int] = Field(default=None, title="Intermediates", description="Size of the intermediate candidate pool when matching image features to descriptive tags. Higher values may improve quality but are slower.")
num_beams: Optional[int] = Field(default=None, title="Num Beams", description="Number of beams for beam search during caption generation. Higher values search more possibilities but are slower.")
max_length: int | None = Field(default=None, title="Max Length", description="Maximum number of tokens in the generated caption.")
chunk_size: int | None = Field(default=None, title="Chunk Size", description="Batch size for processing description candidates (flavors). Higher values speed up captioning but increase VRAM usage.")
min_flavors: int | None = Field(default=None, title="Min Flavors", description="Minimum number of descriptive tags (flavors) to keep in the final prompt.")
max_flavors: int | None = Field(default=None, title="Max Flavors", description="Maximum number of descriptive tags (flavors) to keep in the final prompt.")
flavor_count: int | None = Field(default=None, title="Intermediates", description="Size of the intermediate candidate pool when matching image features to descriptive tags. Higher values may improve quality but are slower.")
num_beams: int | None = Field(default=None, title="Num Beams", description="Number of beams for beam search during caption generation. Higher values search more possibilities but are slower.")
class ResCaption(BaseModel):
"""Response model for image captioning results."""
caption: Optional[str] = Field(default=None, title="Caption", description="Generated caption/prompt describing the image content and style.")
medium: Optional[str] = Field(default=None, title="Medium", description="Detected artistic medium (e.g., 'oil painting', 'digital art', 'photograph'). Only returned when analyze=True.")
artist: Optional[str] = Field(default=None, title="Artist", description="Detected similar artist style (e.g., 'by greg rutkowski'). Only returned when analyze=True.")
movement: Optional[str] = Field(default=None, title="Movement", description="Detected art movement (e.g., 'art nouveau', 'impressionism'). Only returned when analyze=True.")
trending: Optional[str] = Field(default=None, title="Trending", description="Trending/platform tags (e.g., 'trending on artstation'). Only returned when analyze=True.")
flavor: Optional[str] = Field(default=None, title="Flavor", description="Additional descriptive elements (e.g., 'cinematic lighting', 'highly detailed'). Only returned when analyze=True.")
caption: str | None = Field(default=None, title="Caption", description="Generated caption/prompt describing the image content and style.")
medium: str | None = Field(default=None, title="Medium", description="Detected artistic medium (e.g., 'oil painting', 'digital art', 'photograph'). Only returned when analyze=True.")
artist: str | None = Field(default=None, title="Artist", description="Detected similar artist style (e.g., 'by greg rutkowski'). Only returned when analyze=True.")
movement: str | None = Field(default=None, title="Movement", description="Detected art movement (e.g., 'art nouveau', 'impressionism'). Only returned when analyze=True.")
trending: str | None = Field(default=None, title="Trending", description="Trending/platform tags (e.g., 'trending on artstation'). Only returned when analyze=True.")
flavor: str | None = Field(default=None, title="Flavor", description="Additional descriptive elements (e.g., 'cinematic lighting', 'highly detailed'). Only returned when analyze=True.")
class ReqVQA(BaseModel):
"""Request model for Vision-Language Model (VLM) captioning.
@@ -74,32 +74,32 @@ class ReqVQA(BaseModel):
image: str = Field(default="", title="Image", description="Image to caption. Must be a Base64 encoded string containing the image data.")
model: str = Field(default="Alibaba Qwen 2.5 VL 3B", title="Model", description="Select which model to use for Visual Language tasks. Use GET /sdapi/v1/vqa/models for full list. Models which support thinking mode are indicated in capabilities.")
question: str = Field(default="describe the image", title="Question/Task", description="Task for the model to perform. Common tasks: 'Short Caption', 'Normal Caption', 'Long Caption'. Set to 'Use Prompt' to pass custom text via the prompt field. Florence-2 tasks: 'Object Detection', 'OCR (Read Text)', 'Phrase Grounding', 'Dense Region Caption', 'Region Proposal', 'OCR with Regions'. PromptGen tasks: 'Analyze', 'Generate Tags', 'Mixed Caption'. Moondream tasks: 'Point at...', 'Detect all...', 'Detect Gaze' (Moondream 2 only). Use GET /sdapi/v1/vqa/prompts?model=<name> to list tasks available for a specific model.")
prompt: Optional[str] = Field(default=None, title="Prompt", description="Custom prompt text. Required when question is 'Use Prompt'. For 'Point at...' tasks, specify what to find (e.g., 'the red car'). For 'Detect all...' tasks, specify what to detect (e.g., 'faces').")
prompt: str | None = Field(default=None, title="Prompt", description="Custom prompt text. Required when question is 'Use Prompt'. For 'Point at...' tasks, specify what to find (e.g., 'the red car'). For 'Detect all...' tasks, specify what to detect (e.g., 'faces').")
system: str = Field(default="You are image captioning expert, creative, unbiased and uncensored.", title="System Prompt", description="System prompt controls behavior of the LLM. Processed first and persists throughout conversation. Has highest priority weighting and is always appended at the beginning of the sequence. Use for: Response formatting rules, role definition, style.")
include_annotated: bool = Field(default=False, title="Include Annotated Image", description="If True and the task produces detection results (object detection, point detection, gaze), returns annotated image with bounding boxes/points drawn. Only applicable for detection tasks on models like Florence-2 and Moondream.")
# LLM generation parameters (optional overrides)
max_tokens: Optional[int] = Field(default=None, title="Max Tokens", description="Maximum number of tokens the model can generate in its response. The model is not aware of this limit during generation; it simply sets the hard limit for the length and will forcefully cut off the response when reached.")
temperature: Optional[float] = Field(default=None, title="Temperature", description="Controls randomness in token selection. Lower values (e.g., 0.1) make outputs more focused and deterministic, always choosing high-probability tokens. Higher values (e.g., 0.9) increase creativity and diversity by allowing less probable tokens. Set to 0 for fully deterministic output.")
top_k: Optional[int] = Field(default=None, title="Top-K", description="Limits token selection to the K most likely candidates at each step. Lower values (e.g., 40) make outputs more focused and predictable, while higher values allow more diverse choices. Set to 0 to disable.")
top_p: Optional[float] = Field(default=None, title="Top-P", description="Selects tokens from the smallest set whose cumulative probability exceeds P (e.g., 0.9). Dynamically adapts the number of candidates based on model confidence; fewer options when certain, more when uncertain. Set to 1 to disable.")
num_beams: Optional[int] = Field(default=None, title="Num Beams", description="Maintains multiple candidate paths simultaneously and selects the overall best sequence. More thorough but much slower and less creative than random sampling. Generally not recommended; most modern VLMs perform better with sampling methods. Set to 1 to disable.")
do_sample: Optional[bool] = Field(default=None, title="Use Samplers", description="Enable to use sampling (randomly selecting tokens based on sampling methods like Top-K or Top-P) or disable to use greedy decoding (selecting the most probable token at each step). Enabling makes outputs more diverse and creative but less deterministic.")
thinking_mode: Optional[bool] = Field(default=None, title="Thinking Mode", description="Enables thinking/reasoning, allowing the model to take more time to generate responses. Can lead to more thoughtful and detailed answers but increases response time. Only works with models that support this feature.")
prefill: Optional[str] = Field(default=None, title="Prefill Text", description="Pre-fills the start of the model's response to guide its output format or content by forcing it to continue the prefill text. Prefill is filtered out and does not appear in the final response unless keep_prefill is True. Leave empty to let the model generate from scratch.")
keep_thinking: Optional[bool] = Field(default=None, title="Keep Thinking Trace", description="Include the model's reasoning process in the final output. Useful for understanding how the model arrived at its answer. Only works with models that support thinking mode.")
keep_prefill: Optional[bool] = Field(default=None, title="Keep Prefill", description="Include the prefill text at the beginning of the final output. If disabled, the prefill text used to guide the model is removed from the result.")
max_tokens: int | None = Field(default=None, title="Max Tokens", description="Maximum number of tokens the model can generate in its response. The model is not aware of this limit during generation; it simply sets the hard limit for the length and will forcefully cut off the response when reached.")
temperature: float | None = Field(default=None, title="Temperature", description="Controls randomness in token selection. Lower values (e.g., 0.1) make outputs more focused and deterministic, always choosing high-probability tokens. Higher values (e.g., 0.9) increase creativity and diversity by allowing less probable tokens. Set to 0 for fully deterministic output.")
top_k: int | None = Field(default=None, title="Top-K", description="Limits token selection to the K most likely candidates at each step. Lower values (e.g., 40) make outputs more focused and predictable, while higher values allow more diverse choices. Set to 0 to disable.")
top_p: float | None = Field(default=None, title="Top-P", description="Selects tokens from the smallest set whose cumulative probability exceeds P (e.g., 0.9). Dynamically adapts the number of candidates based on model confidence; fewer options when certain, more when uncertain. Set to 1 to disable.")
num_beams: int | None = Field(default=None, title="Num Beams", description="Maintains multiple candidate paths simultaneously and selects the overall best sequence. More thorough but much slower and less creative than random sampling. Generally not recommended; most modern VLMs perform better with sampling methods. Set to 1 to disable.")
do_sample: bool | None = Field(default=None, title="Use Samplers", description="Enable to use sampling (randomly selecting tokens based on sampling methods like Top-K or Top-P) or disable to use greedy decoding (selecting the most probable token at each step). Enabling makes outputs more diverse and creative but less deterministic.")
thinking_mode: bool | None = Field(default=None, title="Thinking Mode", description="Enables thinking/reasoning, allowing the model to take more time to generate responses. Can lead to more thoughtful and detailed answers but increases response time. Only works with models that support this feature.")
prefill: str | None = Field(default=None, title="Prefill Text", description="Pre-fills the start of the model's response to guide its output format or content by forcing it to continue the prefill text. Prefill is filtered out and does not appear in the final response unless keep_prefill is True. Leave empty to let the model generate from scratch.")
keep_thinking: bool | None = Field(default=None, title="Keep Thinking Trace", description="Include the model's reasoning process in the final output. Useful for understanding how the model arrived at its answer. Only works with models that support thinking mode.")
keep_prefill: bool | None = Field(default=None, title="Keep Prefill", description="Include the prefill text at the beginning of the final output. If disabled, the prefill text used to guide the model is removed from the result.")
class ResVQA(BaseModel):
"""Response model for VLM captioning results."""
answer: Optional[str] = Field(default=None, title="Answer", description="Generated caption, answer, or analysis from the VLM. Format depends on the question/task type.")
annotated_image: Optional[str] = Field(default=None, title="Annotated Image", description="Base64 encoded PNG image with detection results drawn (bounding boxes, points). Only returned when include_annotated=True and the task produces detection results.")
answer: str | None = Field(default=None, title="Answer", description="Generated caption, answer, or analysis from the VLM. Format depends on the question/task type.")
annotated_image: str | None = Field(default=None, title="Annotated Image", description="Base64 encoded PNG image with detection results drawn (bounding boxes, points). Only returned when include_annotated=True and the task produces detection results.")
class ItemVLMModel(BaseModel):
"""VLM model information."""
name: str = Field(title="Name", description="Display name of the model")
repo: str = Field(title="Repository", description="HuggingFace repository ID")
prompts: List[str] = Field(title="Prompts", description="Available prompts/tasks for this model")
capabilities: List[str] = Field(title="Capabilities", description="Model capabilities. Possible values: 'caption' (image captioning), 'vqa' (visual question answering), 'detection' (object/point detection), 'ocr' (text recognition), 'thinking' (reasoning mode support).")
prompts: list[str] = Field(title="Prompts", description="Available prompts/tasks for this model")
capabilities: list[str] = Field(title="Capabilities", description="Model capabilities. Possible values: 'caption' (image captioning), 'vqa' (visual question answering), 'detection' (object/point detection), 'ocr' (text recognition), 'thinking' (reasoning mode support).")
class ResVLMPrompts(BaseModel):
"""Available VLM prompts grouped by category.
@@ -107,12 +107,12 @@ class ResVLMPrompts(BaseModel):
When called without ``model`` parameter, returns all prompt categories.
When called with ``model``, returns only the ``available`` field with prompts for that model.
"""
common: Optional[List[str]] = Field(default=None, title="Common", description="Prompts available for all models: Use Prompt, Short/Normal/Long Caption.")
florence: Optional[List[str]] = Field(default=None, title="Florence", description="Florence-2 base model tasks: Phrase Grounding, Object Detection, Dense Region Caption, Region Proposal, OCR (Read Text), OCR with Regions.")
promptgen: Optional[List[str]] = Field(default=None, title="PromptGen", description="MiaoshouAI PromptGen fine-tune tasks: Analyze, Generate Tags, Mixed Caption, Mixed Caption+. Only available on PromptGen models.")
moondream: Optional[List[str]] = Field(default=None, title="Moondream", description="Moondream 2 and 3 tasks: Point at..., Detect all...")
moondream2_only: Optional[List[str]] = Field(default=None, title="Moondream 2 Only", description="Moondream 2 exclusive tasks: Detect Gaze. Not available in Moondream 3.")
available: Optional[List[str]] = Field(default=None, title="Available", description="Populated only when filtering by model. Contains the combined list of prompts available for the specified model.")
common: list[str] | None = Field(default=None, title="Common", description="Prompts available for all models: Use Prompt, Short/Normal/Long Caption.")
florence: list[str] | None = Field(default=None, title="Florence", description="Florence-2 base model tasks: Phrase Grounding, Object Detection, Dense Region Caption, Region Proposal, OCR (Read Text), OCR with Regions.")
promptgen: list[str] | None = Field(default=None, title="PromptGen", description="MiaoshouAI PromptGen fine-tune tasks: Analyze, Generate Tags, Mixed Caption, Mixed Caption+. Only available on PromptGen models.")
moondream: list[str] | None = Field(default=None, title="Moondream", description="Moondream 2 and 3 tasks: Point at..., Detect all...")
moondream2_only: list[str] | None = Field(default=None, title="Moondream 2 Only", description="Moondream 2 exclusive tasks: Detect Gaze. Not available in Moondream 3.")
available: list[str] | None = Field(default=None, title="Available", description="Populated only when filtering by model. Contains the combined list of prompts available for the specified model.")
class ItemTaggerModel(BaseModel):
"""Tagger model information."""
@@ -136,7 +136,7 @@ class ReqTagger(BaseModel):
class ResTagger(BaseModel):
"""Response model for image tagging results."""
tags: str = Field(title="Tags", description="Comma-separated list of detected tags")
scores: Optional[dict] = Field(default=None, title="Scores", description="Tag confidence scores (when show_scores=True)")
scores: dict | None = Field(default=None, title="Scores", description="Tag confidence scores (when show_scores=True)")
# =============================================================================
@@ -158,12 +158,12 @@ class ReqCaptionOpenCLIP(BaseModel):
blip_model: str = Field(default="blip-large", title="Caption Model", description="BLIP model used to generate the initial image caption.")
mode: str = Field(default="best", title="Mode", description="Caption mode: 'best' (highest quality, slowest), 'fast' (quick, fewer flavors), 'classic' (balanced), 'caption' (BLIP only, no CLIP matching), 'negative' (for negative prompts).")
analyze: bool = Field(default=False, title="Analyze", description="If True, returns detailed breakdown (medium, artist, movement, trending, flavor).")
max_length: Optional[int] = Field(default=None, title="Max Length", description="Maximum tokens in generated caption.")
chunk_size: Optional[int] = Field(default=None, title="Chunk Size", description="Batch size for processing flavors.")
min_flavors: Optional[int] = Field(default=None, title="Min Flavors", description="Minimum descriptive tags to keep.")
max_flavors: Optional[int] = Field(default=None, title="Max Flavors", description="Maximum descriptive tags to keep.")
flavor_count: Optional[int] = Field(default=None, title="Intermediates", description="Size of intermediate candidate pool.")
num_beams: Optional[int] = Field(default=None, title="Num Beams", description="Beams for beam search during caption generation.")
max_length: int | None = Field(default=None, title="Max Length", description="Maximum tokens in generated caption.")
chunk_size: int | None = Field(default=None, title="Chunk Size", description="Batch size for processing flavors.")
min_flavors: int | None = Field(default=None, title="Min Flavors", description="Minimum descriptive tags to keep.")
max_flavors: int | None = Field(default=None, title="Max Flavors", description="Maximum descriptive tags to keep.")
flavor_count: int | None = Field(default=None, title="Intermediates", description="Size of intermediate candidate pool.")
num_beams: int | None = Field(default=None, title="Num Beams", description="Beams for beam search during caption generation.")
class ReqCaptionTagger(BaseModel):
@@ -196,24 +196,24 @@ class ReqCaptionVLM(BaseModel):
image: str = Field(default="", title="Image", description="Image to caption. Must be a Base64 encoded string.")
model: str = Field(default="Alibaba Qwen 2.5 VL 3B", title="Model", description="VLM model to use. See GET /sdapi/v1/vqa/models for full list.")
question: str = Field(default="describe the image", title="Question/Task", description="Task to perform: 'Short Caption', 'Normal Caption', 'Long Caption', 'Use Prompt' (custom text via prompt field). Model-specific tasks available via GET /sdapi/v1/vqa/prompts.")
prompt: Optional[str] = Field(default=None, title="Prompt", description="Custom prompt text when question is 'Use Prompt'.")
prompt: str | None = Field(default=None, title="Prompt", description="Custom prompt text when question is 'Use Prompt'.")
system: str = Field(default="You are image captioning expert, creative, unbiased and uncensored.", title="System Prompt", description="System prompt for LLM behavior.")
include_annotated: bool = Field(default=False, title="Include Annotated Image", description="Return annotated image for detection tasks.")
max_tokens: Optional[int] = Field(default=None, title="Max Tokens", description="Maximum tokens in response.")
temperature: Optional[float] = Field(default=None, title="Temperature", description="Randomness in token selection (0=deterministic, 0.9=creative).")
top_k: Optional[int] = Field(default=None, title="Top-K", description="Limit to K most likely tokens per step.")
top_p: Optional[float] = Field(default=None, title="Top-P", description="Nucleus sampling threshold.")
num_beams: Optional[int] = Field(default=None, title="Num Beams", description="Beam search width (1=disabled).")
do_sample: Optional[bool] = Field(default=None, title="Use Samplers", description="Enable sampling vs greedy decoding.")
thinking_mode: Optional[bool] = Field(default=None, title="Thinking Mode", description="Enable reasoning mode (supported models only).")
prefill: Optional[str] = Field(default=None, title="Prefill Text", description="Pre-fill response start to guide output.")
keep_thinking: Optional[bool] = Field(default=None, title="Keep Thinking Trace", description="Include reasoning in output.")
keep_prefill: Optional[bool] = Field(default=None, title="Keep Prefill", description="Keep prefill text in final output.")
max_tokens: int | None = Field(default=None, title="Max Tokens", description="Maximum tokens in response.")
temperature: float | None = Field(default=None, title="Temperature", description="Randomness in token selection (0=deterministic, 0.9=creative).")
top_k: int | None = Field(default=None, title="Top-K", description="Limit to K most likely tokens per step.")
top_p: float | None = Field(default=None, title="Top-P", description="Nucleus sampling threshold.")
num_beams: int | None = Field(default=None, title="Num Beams", description="Beam search width (1=disabled).")
do_sample: bool | None = Field(default=None, title="Use Samplers", description="Enable sampling vs greedy decoding.")
thinking_mode: bool | None = Field(default=None, title="Thinking Mode", description="Enable reasoning mode (supported models only).")
prefill: str | None = Field(default=None, title="Prefill Text", description="Pre-fill response start to guide output.")
keep_thinking: bool | None = Field(default=None, title="Keep Thinking Trace", description="Include reasoning in output.")
keep_prefill: bool | None = Field(default=None, title="Keep Prefill", description="Keep prefill text in final output.")
# Discriminated union for the dispatch endpoint
ReqCaptionDispatch = Annotated[
Union[ReqCaptionOpenCLIP, ReqCaptionTagger, ReqCaptionVLM],
ReqCaptionOpenCLIP | ReqCaptionTagger | ReqCaptionVLM,
Field(discriminator="backend")
]
@@ -226,18 +226,18 @@ class ResCaptionDispatch(BaseModel):
# Common
backend: str = Field(title="Backend", description="The backend that processed the request: 'openclip', 'tagger', or 'vlm'.")
# OpenCLIP fields
caption: Optional[str] = Field(default=None, title="Caption", description="Generated caption (OpenCLIP backend).")
medium: Optional[str] = Field(default=None, title="Medium", description="Detected artistic medium (OpenCLIP with analyze=True).")
artist: Optional[str] = Field(default=None, title="Artist", description="Detected artist style (OpenCLIP with analyze=True).")
movement: Optional[str] = Field(default=None, title="Movement", description="Detected art movement (OpenCLIP with analyze=True).")
trending: Optional[str] = Field(default=None, title="Trending", description="Trending tags (OpenCLIP with analyze=True).")
flavor: Optional[str] = Field(default=None, title="Flavor", description="Flavor descriptors (OpenCLIP with analyze=True).")
caption: str | None = Field(default=None, title="Caption", description="Generated caption (OpenCLIP backend).")
medium: str | None = Field(default=None, title="Medium", description="Detected artistic medium (OpenCLIP with analyze=True).")
artist: str | None = Field(default=None, title="Artist", description="Detected artist style (OpenCLIP with analyze=True).")
movement: str | None = Field(default=None, title="Movement", description="Detected art movement (OpenCLIP with analyze=True).")
trending: str | None = Field(default=None, title="Trending", description="Trending tags (OpenCLIP with analyze=True).")
flavor: str | None = Field(default=None, title="Flavor", description="Flavor descriptors (OpenCLIP with analyze=True).")
# Tagger fields
tags: Optional[str] = Field(default=None, title="Tags", description="Comma-separated tags (Tagger backend).")
scores: Optional[dict] = Field(default=None, title="Scores", description="Tag confidence scores (Tagger with show_scores=True).")
tags: str | None = Field(default=None, title="Tags", description="Comma-separated tags (Tagger backend).")
scores: dict | None = Field(default=None, title="Scores", description="Tag confidence scores (Tagger with show_scores=True).")
# VLM fields
answer: Optional[str] = Field(default=None, title="Answer", description="VLM response (VLM backend).")
annotated_image: Optional[str] = Field(default=None, title="Annotated Image", description="Base64 annotated image (VLM with include_annotated=True).")
answer: str | None = Field(default=None, title="Answer", description="VLM response (VLM backend).")
annotated_image: str | None = Field(default=None, title="Annotated Image", description="Base64 annotated image (VLM with include_annotated=True).")
# =============================================================================
@@ -596,7 +596,7 @@ def get_vqa_models():
return models_list
def get_vqa_prompts(model: Optional[str] = None):
def get_vqa_prompts(model: str | None = None):
"""
List available prompts/tasks for VLM models.
@@ -653,11 +653,11 @@ def get_tagger_models():
def register_api():
from modules.shared import api
api.add_api_route("/sdapi/v1/openclip", get_caption, methods=["GET"], response_model=List[str], tags=["Caption"])
api.add_api_route("/sdapi/v1/openclip", get_caption, methods=["GET"], response_model=list[str], tags=["Caption"])
api.add_api_route("/sdapi/v1/caption", post_caption_dispatch, methods=["POST"], response_model=ResCaptionDispatch, tags=["Caption"])
api.add_api_route("/sdapi/v1/openclip", post_caption, methods=["POST"], response_model=ResCaption, tags=["Caption"])
api.add_api_route("/sdapi/v1/vqa", post_vqa, methods=["POST"], response_model=ResVQA, tags=["Caption"])
api.add_api_route("/sdapi/v1/vqa/models", get_vqa_models, methods=["GET"], response_model=List[ItemVLMModel], tags=["Caption"])
api.add_api_route("/sdapi/v1/vqa/models", get_vqa_models, methods=["GET"], response_model=list[ItemVLMModel], tags=["Caption"])
api.add_api_route("/sdapi/v1/vqa/prompts", get_vqa_prompts, methods=["GET"], response_model=ResVLMPrompts, tags=["Caption"])
api.add_api_route("/sdapi/v1/tagger", post_tagger, methods=["POST"], response_model=ResTagger, tags=["Caption"])
api.add_api_route("/sdapi/v1/tagger/models", get_tagger_models, methods=["GET"], response_model=List[ItemTaggerModel], tags=["Caption"])
api.add_api_route("/sdapi/v1/tagger/models", get_tagger_models, methods=["GET"], response_model=list[ItemTaggerModel], tags=["Caption"])
modules/api/control.py
+6 -6
View File
@@ -1,4 +1,4 @@
from typing import Optional, List
from typing import Optional
from threading import Lock
from pydantic import BaseModel, Field # pylint: disable=no-name-in-module
from modules import errors, shared, processing_helpers
@@ -43,9 +43,9 @@ ReqControl = models.create_model_from_signature(
{"key": "send_images", "type": bool, "default": True},
{"key": "save_images", "type": bool, "default": False},
{"key": "alwayson_scripts", "type": dict, "default": {}},
{"key": "ip_adapter", "type": Optional[List[models.ItemIPAdapter]], "default": None, "exclude": True},
{"key": "ip_adapter", "type": Optional[list[models.ItemIPAdapter]], "default": None, "exclude": True},
{"key": "face", "type": Optional[models.ItemFace], "default": None, "exclude": True},
{"key": "control", "type": Optional[List[ItemControl]], "default": [], "exclude": True},
{"key": "control", "type": Optional[list[ItemControl]], "default": [], "exclude": True},
{"key": "xyz", "type": Optional[ItemXYZ], "default": None, "exclude": True},
# {"key": "extra", "type": Optional[dict], "default": {}, "exclude": True},
]
@@ -55,13 +55,13 @@ if not hasattr(ReqControl, "__config__"):
class ResControl(BaseModel):
images: List[str] = Field(default=None, title="Images", description="")
processed: List[str] = Field(default=None, title="Processed", description="")
images: list[str] = Field(default=None, title="Images", description="")
processed: list[str] = Field(default=None, title="Processed", description="")
params: dict = Field(default={}, title="Settings", description="")
info: str = Field(default="", title="Info", description="")
class APIControl():
class APIControl:
def __init__(self, queue_lock: Lock):
self.queue_lock = queue_lock
self.default_script_arg = []
modules/api/endpoints.py
+2 -3
View File
@@ -1,4 +1,3 @@
from typing import Optional
from modules import shared
from modules.api import models, helpers
@@ -43,7 +42,7 @@ def get_sd_models():
checkpoints.append(model)
return checkpoints
def get_controlnets(model_type: Optional[str] = None):
def get_controlnets(model_type: str | None = None):
from modules.control.units.controlnet import api_list_models
return api_list_models(model_type)
@@ -60,7 +59,7 @@ def get_embeddings():
return models.ResEmbeddings(loaded=[], skipped=[])
return models.ResEmbeddings(loaded=list(db.word_embeddings.keys()), skipped=list(db.skipped_embeddings.keys()))
def get_extra_networks(page: Optional[str] = None, name: Optional[str] = None, filename: Optional[str] = None, title: Optional[str] = None, fullname: Optional[str] = None, hash: Optional[str] = None): # pylint: disable=redefined-builtin
def get_extra_networks(page: str | None = None, name: str | None = None, filename: str | None = None, title: str | None = None, fullname: str | None = None, hash: str | None = None): # pylint: disable=redefined-builtin
res = []
for pg in shared.extra_networks:
if page is not None and pg.name != page.lower():
modules/api/gallery.py
+3 -4
View File
@@ -2,7 +2,6 @@ import io
import os
import time
import base64
from typing import List, Union
from urllib.parse import quote, unquote
from fastapi import FastAPI
from fastapi.responses import JSONResponse
@@ -52,7 +51,7 @@ class ConnectionManager:
debug(f'Browser WS disconnect: client={ws.client.host}')
self.active.remove(ws)
async def send(self, ws: WebSocket, data: Union[str, dict, bytes]):
async def send(self, ws: WebSocket, data: str | dict | bytes):
# debug(f'Browser WS send: client={ws.client.host} data={type(data)}')
if ws.client_state != WebSocketState.CONNECTED:
return
@@ -65,7 +64,7 @@ class ConnectionManager:
else:
debug(f'Browser WS send: client={ws.client.host} data={type(data)} unknown')
async def broadcast(self, data: Union[str, dict, bytes]):
async def broadcast(self, data: str | dict | bytes):
for ws in self.active:
await self.send(ws, data)
@@ -206,7 +205,7 @@ def register_api(app: FastAPI): # register api
shared.log.error(f'Gallery: {folder} {e}')
return []
shared.api.add_api_route("/sdapi/v1/browser/folders", get_folders, methods=["GET"], response_model=List[str])
shared.api.add_api_route("/sdapi/v1/browser/folders", get_folders, methods=["GET"], response_model=list[str])
shared.api.add_api_route("/sdapi/v1/browser/thumb", get_thumb, methods=["GET"], response_model=dict)
shared.api.add_api_route("/sdapi/v1/browser/files", ht_files, methods=["GET"], response_model=list)
modules/api/generate.py
+1 -1
View File
@@ -9,7 +9,7 @@ from modules.paths import resolve_output_path
errors.install()
class APIGenerate():
class APIGenerate:
def __init__(self, queue_lock: Lock):
self.queue_lock = queue_lock
self.default_script_arg_txt2img = []
modules/api/loras.py
+1 -2
View File
@@ -1,4 +1,3 @@
from typing import List
from fastapi.exceptions import HTTPException
@@ -25,5 +24,5 @@ def post_refresh_loras():
def register_api():
from modules.shared import api
api.add_api_route("/sdapi/v1/lora", get_lora, methods=["GET"], response_model=dict)
api.add_api_route("/sdapi/v1/loras", get_loras, methods=["GET"], response_model=List[dict])
api.add_api_route("/sdapi/v1/loras", get_loras, methods=["GET"], response_model=list[dict])
api.add_api_route("/sdapi/v1/refresh-loras", post_refresh_loras, methods=["POST"])
modules/api/models.py
+120 -82
View File
@@ -1,7 +1,15 @@
import re
import inspect
from typing import Any, Optional, Dict, List, Type, Callable, Union
from pydantic import BaseModel, Field, create_model # pylint: disable=no-name-in-module
from typing import Any, Optional, Union
from collections.abc import Callable
import pydantic
from pydantic import BaseModel, Field, create_model
try:
from pydantic import ConfigDict
PYDANTIC_V2 = True
except ImportError:
ConfigDict = None
PYDANTIC_V2 = False
from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img
import modules.shared as shared
@@ -41,8 +49,10 @@ class PydanticModelGenerator:
model_name: str = None,
class_instance = None,
additional_fields = None,
exclude_fields: List = [],
exclude_fields: list = None,
):
if exclude_fields is None:
exclude_fields = []
def field_type_generator(_k, v):
field_type = v.annotation
return Optional[field_type]
@@ -80,12 +90,15 @@ class PydanticModelGenerator:
def generate_model(self):
model_fields = { d.field: (d.field_type, Field(default=d.field_value, alias=d.field_alias, exclude=d.field_exclude)) for d in self._model_def }
DynamicModel = create_model(self._model_name, **model_fields)
try:
DynamicModel.__config__.allow_population_by_field_name = True
DynamicModel.__config__.allow_mutation = True
except Exception:
pass
if PYDANTIC_V2:
config = ConfigDict(arbitrary_types_allowed=True, from_attributes=True, populate_by_name=True)
else:
class Config:
arbitrary_types_allowed = True
orm_mode = True
allow_population_by_field_name = True
config = Config
DynamicModel = create_model(self._model_name, __config__=config, **model_fields)
return DynamicModel
### item classes
@@ -100,49 +113,49 @@ class ItemVae(BaseModel):
class ItemUpscaler(BaseModel):
name: str = Field(title="Name")
model_name: Optional[str] = Field(title="Model Name")
model_path: Optional[str] = Field(title="Path")
model_url: Optional[str] = Field(title="URL")
scale: Optional[float] = Field(title="Scale")
model_name: str | None = Field(title="Model Name")
model_path: str | None = Field(title="Path")
model_url: str | None = Field(title="URL")
scale: float | None = Field(title="Scale")
class ItemModel(BaseModel):
title: str = Field(title="Title")
model_name: str = Field(title="Model Name")
filename: str = Field(title="Filename")
type: str = Field(title="Model type")
sha256: Optional[str] = Field(title="SHA256 hash")
hash: Optional[str] = Field(title="Short hash")
config: Optional[str] = Field(title="Config file")
sha256: str | None = Field(title="SHA256 hash")
hash: str | None = Field(title="Short hash")
config: str | None = Field(title="Config file")
class ItemHypernetwork(BaseModel):
name: str = Field(title="Name")
path: Optional[str] = Field(title="Path")
path: str | None = Field(title="Path")
class ItemDetailer(BaseModel):
name: str = Field(title="Name")
path: Optional[str] = Field(title="Path")
path: str | None = Field(title="Path")
class ItemGAN(BaseModel):
name: str = Field(title="Name")
path: Optional[str] = Field(title="Path")
scale: Optional[int] = Field(title="Scale")
path: str | None = Field(title="Path")
scale: int | None = Field(title="Scale")
class ItemStyle(BaseModel):
name: str = Field(title="Name")
prompt: Optional[str] = Field(title="Prompt")
negative_prompt: Optional[str] = Field(title="Negative Prompt")
extra: Optional[str] = Field(title="Extra")
filename: Optional[str] = Field(title="Filename")
preview: Optional[str] = Field(title="Preview")
prompt: str | None = Field(title="Prompt")
negative_prompt: str | None = Field(title="Negative Prompt")
extra: str | None = Field(title="Extra")
filename: str | None = Field(title="Filename")
preview: str | None = Field(title="Preview")
class ItemExtraNetwork(BaseModel):
name: str = Field(title="Name")
type: str = Field(title="Type")
title: Optional[str] = Field(title="Title")
fullname: Optional[str] = Field(title="Fullname")
filename: Optional[str] = Field(title="Filename")
hash: Optional[str] = Field(title="Hash")
preview: Optional[str] = Field(title="Preview image URL")
title: str | None = Field(title="Title")
fullname: str | None = Field(title="Fullname")
filename: str | None = Field(title="Filename")
hash: str | None = Field(title="Hash")
preview: str | None = Field(title="Preview image URL")
class ItemArtist(BaseModel):
name: str = Field(title="Name")
@@ -150,16 +163,16 @@ class ItemArtist(BaseModel):
category: str = Field(title="Category")
class ItemEmbedding(BaseModel):
step: Optional[int] = Field(title="Step", description="The number of steps that were used to train this embedding, if available")
sd_checkpoint: Optional[str] = Field(title="SD Checkpoint", description="The hash of the checkpoint this embedding was trained on, if available")
sd_checkpoint_name: Optional[str] = Field(title="SD Checkpoint Name", description="The name of the checkpoint this embedding was trained on, if available. Note that this is the name that was used by the trainer; for a stable identifier, use `sd_checkpoint` instead")
step: int | None = Field(title="Step", description="The number of steps that were used to train this embedding, if available")
sd_checkpoint: str | None = Field(title="SD Checkpoint", description="The hash of the checkpoint this embedding was trained on, if available")
sd_checkpoint_name: str | None = Field(title="SD Checkpoint Name", description="The name of the checkpoint this embedding was trained on, if available. Note that this is the name that was used by the trainer; for a stable identifier, use `sd_checkpoint` instead")
shape: int = Field(title="Shape", description="The length of each individual vector in the embedding")
vectors: int = Field(title="Vectors", description="The number of vectors in the embedding")
class ItemIPAdapter(BaseModel):
adapter: str = Field(title="Adapter", default="Base", description="IP adapter name")
images: List[str] = Field(title="Image", default=[], description="IP adapter input images")
masks: Optional[List[str]] = Field(title="Mask", default=[], description="IP adapter mask images")
images: list[str] = Field(title="Image", default=[], description="IP adapter input images")
masks: list[str] | None = Field(title="Mask", default=[], description="IP adapter mask images")
scale: float = Field(title="Scale", default=0.5, ge=0, le=1, description="IP adapter scale")
start: float = Field(title="Start", default=0.0, ge=0, le=1, description="IP adapter start step")
end: float = Field(title="End", default=1.0, gt=0, le=1, description="IP adapter end step")
@@ -183,17 +196,17 @@ class ItemFace(BaseModel):
class ScriptArg(BaseModel):
label: str = Field(default=None, title="Label", description="Name of the argument in UI")
value: Optional[Any] = Field(default=None, title="Value", description="Default value of the argument")
minimum: Optional[Any] = Field(default=None, title="Minimum", description="Minimum allowed value for the argumentin UI")
maximum: Optional[Any] = Field(default=None, title="Minimum", description="Maximum allowed value for the argumentin UI")
step: Optional[Any] = Field(default=None, title="Minimum", description="Step for changing value of the argumentin UI")
choices: Optional[Any] = Field(default=None, title="Choices", description="Possible values for the argument")
value: Any | None = Field(default=None, title="Value", description="Default value of the argument")
minimum: Any | None = Field(default=None, title="Minimum", description="Minimum allowed value for the argumentin UI")
maximum: Any | None = Field(default=None, title="Minimum", description="Maximum allowed value for the argumentin UI")
step: Any | None = Field(default=None, title="Minimum", description="Step for changing value of the argumentin UI")
choices: Any | None = Field(default=None, title="Choices", description="Possible values for the argument")
class ItemScript(BaseModel):
name: str = Field(default=None, title="Name", description="Script name")
is_alwayson: bool = Field(default=None, title="IsAlwayson", description="Flag specifying whether this script is an alwayson script")
is_img2img: bool = Field(default=None, title="IsImg2img", description="Flag specifying whether this script is an img2img script")
args: List[ScriptArg] = Field(title="Arguments", description="List of script's arguments")
args: list[ScriptArg] = Field(title="Arguments", description="List of script's arguments")
class ItemExtension(BaseModel):
name: str = Field(title="Name", description="Extension name")
@@ -201,13 +214,13 @@ class ItemExtension(BaseModel):
branch: str = Field(default="uknnown", title="Branch", description="Extension Repository Branch")
commit_hash: str = Field(title="Commit Hash", description="Extension Repository Commit Hash")
version: str = Field(title="Version", description="Extension Version")
commit_date: Union[str, int] = Field(title="Commit Date", description="Extension Repository Commit Date")
commit_date: str | int = Field(title="Commit Date", description="Extension Repository Commit Date")
enabled: bool = Field(title="Enabled", description="Flag specifying whether this extension is enabled")
class ItemScheduler(BaseModel):
name: str = Field(title="Name", description="Scheduler name")
cls: str = Field(title="Class", description="Scheduler class name")
options: Dict[str, Any] = Field(title="Options", description="Dictionary of scheduler options")
options: dict[str, Any] = Field(title="Options", description="Dictionary of scheduler options")
### request/response classes
@@ -223,7 +236,7 @@ ReqTxt2Img = PydanticModelGenerator(
{"key": "send_images", "type": bool, "default": True},
{"key": "save_images", "type": bool, "default": False},
{"key": "alwayson_scripts", "type": dict, "default": {}},
{"key": "ip_adapter", "type": Optional[List[ItemIPAdapter]], "default": None, "exclude": True},
{"key": "ip_adapter", "type": Optional[list[ItemIPAdapter]], "default": None, "exclude": True},
{"key": "face", "type": Optional[ItemFace], "default": None, "exclude": True},
{"key": "extra", "type": Optional[dict], "default": {}, "exclude": True},
]
@@ -233,7 +246,7 @@ if not hasattr(ReqTxt2Img, "__config__"):
StableDiffusionTxt2ImgProcessingAPI = ReqTxt2Img
class ResTxt2Img(BaseModel):
images: List[str] = Field(default=None, title="Image", description="The generated images in base64 format.")
images: list[str] = Field(default=None, title="Image", description="The generated images in base64 format.")
parameters: dict
info: str
@@ -253,7 +266,7 @@ ReqImg2Img = PydanticModelGenerator(
{"key": "send_images", "type": bool, "default": True},
{"key": "save_images", "type": bool, "default": False},
{"key": "alwayson_scripts", "type": dict, "default": {}},
{"key": "ip_adapter", "type": Optional[List[ItemIPAdapter]], "default": None, "exclude": True},
{"key": "ip_adapter", "type": Optional[list[ItemIPAdapter]], "default": None, "exclude": True},
{"key": "face_id", "type": Optional[ItemFace], "default": None, "exclude": True},
{"key": "extra", "type": Optional[dict], "default": {}, "exclude": True},
]
@@ -263,7 +276,7 @@ if not hasattr(ReqImg2Img, "__config__"):
StableDiffusionImg2ImgProcessingAPI = ReqImg2Img
class ResImg2Img(BaseModel):
images: List[str] = Field(default=None, title="Image", description="The generated images in base64 format.")
images: list[str] = Field(default=None, title="Image", description="The generated images in base64 format.")
parameters: dict
info: str
@@ -289,9 +302,9 @@ class ResProcess(BaseModel):
class ReqPromptEnhance(BaseModel):
prompt: str = Field(title="Prompt", description="Prompt to enhance")
type: str = Field(title="Type", default='text', description="Type of enhancement to perform")
model: Optional[str] = Field(title="Model", default=None, description="Model to use for enhancement")
system_prompt: Optional[str] = Field(title="System prompt", default=None, description="Model system prompt")
image: Optional[str] = Field(title="Image", default=None, description="Image to work on, must be a Base64 string containing the image's data.")
model: str | None = Field(title="Model", default=None, description="Model to use for enhancement")
system_prompt: str | None = Field(title="System prompt", default=None, description="Model system prompt")
image: str | None = Field(title="Image", default=None, description="Image to work on, must be a Base64 string containing the image's data.")
seed: int = Field(title="Seed", default=-1, description="Seed used to generate the prompt")
nsfw: bool = Field(title="NSFW", default=True, description="Should NSFW content be allowed?")
@@ -306,10 +319,10 @@ class ResProcessImage(ResProcess):
image: str = Field(default=None, title="Image", description="The generated image in base64 format.")
class ReqProcessBatch(ReqProcess):
imageList: List[FileData] = Field(title="Images", description="List of images to work on. Must be Base64 strings")
imageList: list[FileData] = Field(title="Images", description="List of images to work on. Must be Base64 strings")
class ResProcessBatch(ResProcess):
images: List[str] = Field(title="Images", description="The generated images in base64 format.")
images: list[str] = Field(title="Images", description="The generated images in base64 format.")
class ReqImageInfo(BaseModel):
image: str = Field(title="Image", description="The base64 encoded image")
@@ -325,38 +338,38 @@ class ReqGetLog(BaseModel):
class ReqPostLog(BaseModel):
message: Optional[str] = Field(default=None, title="Message", description="The info message to log")
debug: Optional[str] = Field(default=None, title="Debug message", description="The debug message to log")
error: Optional[str] = Field(default=None, title="Error message", description="The error message to log")
message: str | None = Field(default=None, title="Message", description="The info message to log")
debug: str | None = Field(default=None, title="Debug message", description="The debug message to log")
error: str | None = Field(default=None, title="Error message", description="The error message to log")
class ReqHistory(BaseModel):
id: Union[int, str, None] = Field(default=None, title="Task ID", description="Task ID")
id: int | str | None = Field(default=None, title="Task ID", description="Task ID")
class ReqProgress(BaseModel):
skip_current_image: bool = Field(default=False, title="Skip current image", description="Skip current image serialization")
class ResProgress(BaseModel):
id: Union[int, str, None] = Field(title="TaskID", description="Task ID")
id: int | str | None = Field(title="TaskID", description="Task ID")
progress: float = Field(title="Progress", description="The progress with a range of 0 to 1")
eta_relative: float = Field(title="ETA in secs")
state: dict = Field(title="State", description="The current state snapshot")
current_image: Optional[str] = Field(default=None, title="Current image", description="The current image in base64 format. opts.show_progress_every_n_steps is required for this to work.")
textinfo: Optional[str] = Field(default=None, title="Info text", description="Info text used by WebUI.")
current_image: str | None = Field(default=None, title="Current image", description="The current image in base64 format. opts.show_progress_every_n_steps is required for this to work.")
textinfo: str | None = Field(default=None, title="Info text", description="Info text used by WebUI.")
class ResHistory(BaseModel):
id: Union[int, str, None] = Field(title="ID", description="Task ID")
id: int | str | None = Field(title="ID", description="Task ID")
job: str = Field(title="Job", description="Job name")
op: str = Field(title="Operation", description="Job state")
timestamp: Union[float, None] = Field(title="Timestamp", description="Job timestamp")
duration: Union[float, None] = Field(title="Duration", description="Job duration")
outputs: List[str] = Field(title="Outputs", description="List of filenames")
timestamp: float | None = Field(title="Timestamp", description="Job timestamp")
duration: float | None = Field(title="Duration", description="Job duration")
outputs: list[str] = Field(title="Outputs", description="List of filenames")
class ResStatus(BaseModel):
status: str = Field(title="Status", description="Current status")
task: str = Field(title="Task", description="Current job")
timestamp: Optional[str] = Field(title="Timestamp", description="Timestamp of the current job")
timestamp: str | None = Field(title="Timestamp", description="Timestamp of the current job")
current: str = Field(title="Task", description="Current job")
id: Union[int, str, None] = Field(title="ID", description="ID of the current task")
id: int | str | None = Field(title="ID", description="ID of the current task")
job: int = Field(title="Job", description="Current job")
jobs: int = Field(title="Jobs", description="Total jobs")
total: int = Field(title="Total Jobs", description="Total jobs")
@@ -364,9 +377,9 @@ class ResStatus(BaseModel):
steps: int = Field(title="Steps", description="Total steps")
queued: int = Field(title="Queued", description="Number of queued tasks")
uptime: int = Field(title="Uptime", description="Uptime of the server")
elapsed: Optional[float] = Field(default=None, title="Elapsed time")
eta: Optional[float] = Field(default=None, title="ETA in secs")
progress: Optional[float] = Field(default=None, title="Progress", description="The progress with a range of 0 to 1")
elapsed: float | None = Field(default=None, title="Elapsed time")
eta: float | None = Field(default=None, title="ETA in secs")
progress: float | None = Field(default=None, title="Progress", description="The progress with a range of 0 to 1")
class ReqLatentHistory(BaseModel):
name: str = Field(title="Name", description="Name of the history item to select")
@@ -392,7 +405,15 @@ for key, metadata in shared.opts.data_labels.items():
else:
fields.update({key: (Optional[optType], Field())})
OptionsModel = create_model("Options", **fields)
if PYDANTIC_V2:
config = ConfigDict(arbitrary_types_allowed=True, from_attributes=True, populate_by_name=True)
else:
class Config:
arbitrary_types_allowed = True
orm_mode = True
allow_population_by_field_name = True
config = Config
OptionsModel = create_model("Options", __config__=config, **fields)
flags = {}
_options = vars(shared.parser)['_option_string_actions']
@@ -404,7 +425,15 @@ for key in _options:
_type = type(_options[key].default)
flags.update({flag.dest: (_type, Field(default=flag.default, description=flag.help))})
FlagsModel = create_model("Flags", **flags)
if PYDANTIC_V2:
config = ConfigDict(arbitrary_types_allowed=True, from_attributes=True, populate_by_name=True)
else:
class Config:
arbitrary_types_allowed = True
orm_mode = True
allow_population_by_field_name = True
config = Config
FlagsModel = create_model("Flags", __config__=config, **flags)
class ResEmbeddings(BaseModel):
loaded: list = Field(default=None, title="loaded", description="List of loaded embeddings")
@@ -426,9 +455,13 @@ class ResGPU(BaseModel): # definition of http response
# helper function
def create_model_from_signature(func: Callable, model_name: str, base_model: Type[BaseModel] = BaseModel, additional_fields: List = [], exclude_fields: List[str] = []) -> type[BaseModel]:
def create_model_from_signature(func: Callable, model_name: str, base_model: type[BaseModel] = BaseModel, additional_fields: list = None, exclude_fields: list[str] = None) -> type[BaseModel]:
from PIL import Image
if exclude_fields is None:
exclude_fields = []
if additional_fields is None:
additional_fields = []
class Config:
extra = 'allow'
@@ -443,13 +476,13 @@ def create_model_from_signature(func: Callable, model_name: str, base_model: Typ
defaults = (...,) * non_default_args + defaults
keyword_only_params = {param: kwonlydefaults.get(param, Any) for param in kwonlyargs}
for k, v in annotations.items():
if v == List[Image.Image]:
annotations[k] = List[str]
if v == list[Image.Image]:
annotations[k] = list[str]
elif v == Image.Image:
annotations[k] = str
elif str(v) == 'typing.List[modules.control.unit.Unit]':
annotations[k] = List[str]
model_fields = {param: (annotations.get(param, Any), default) for param, default in zip(args, defaults)}
annotations[k] = list[str]
model_fields = {param: (annotations.get(param, Any), default) for param, default in zip(args, defaults, strict=False)}
for fld in additional_fields:
model_def = ModelDef(
@@ -464,16 +497,21 @@ def create_model_from_signature(func: Callable, model_name: str, base_model: Typ
if fld in model_fields:
del model_fields[fld]
if PYDANTIC_V2:
config = ConfigDict(arbitrary_types_allowed=True, from_attributes=True, populate_by_name=True, extra='allow' if varkw else 'ignore')
else:
class Config:
arbitrary_types_allowed = True
orm_mode = True
allow_population_by_field_name = True
extra = 'allow' if varkw else 'ignore'
config = Config
model = create_model(
model_name,
**model_fields,
**keyword_only_params,
__base__=base_model,
__config__=config,
**model_fields,
**keyword_only_params,
)
try:
model.__config__.allow_population_by_field_name = True
model.__config__.allow_mutation = True
except Exception:
pass
return model
modules/api/process.py
+14 -15
View File
@@ -1,4 +1,3 @@
from typing import Optional, List
from threading import Lock
from pydantic import BaseModel, Field # pylint: disable=no-name-in-module
from fastapi.responses import JSONResponse
@@ -15,7 +14,7 @@ errors.install()
class ReqPreprocess(BaseModel):
image: str = Field(title="Image", description="The base64 encoded image")
model: str = Field(title="Model", description="The model to use for preprocessing")
params: Optional[dict] = Field(default={}, title="Settings", description="Preprocessor settings")
params: dict | None = Field(default={}, title="Settings", description="Preprocessor settings")
class ResPreprocess(BaseModel):
model: str = Field(default='', title="Model", description="The processor model used")
@@ -24,20 +23,20 @@ class ResPreprocess(BaseModel):
class ReqMask(BaseModel):
image: str = Field(title="Image", description="The base64 encoded image")
type: str = Field(title="Mask type", description="Type of masking image to return")
mask: Optional[str] = Field(title="Mask", description="If optional maks image is not provided auto-masking will be performed")
model: Optional[str] = Field(title="Model", description="The model to use for preprocessing")
params: Optional[dict] = Field(default={}, title="Settings", description="Preprocessor settings")
mask: str | None = Field(title="Mask", description="If optional maks image is not provided auto-masking will be performed")
model: str | None = Field(title="Model", description="The model to use for preprocessing")
params: dict | None = Field(default={}, title="Settings", description="Preprocessor settings")
class ReqFace(BaseModel):
image: str = Field(title="Image", description="The base64 encoded image")
model: Optional[str] = Field(title="Model", description="The model to use for detection")
model: str | None = Field(title="Model", description="The model to use for detection")
class ResFace(BaseModel):
classes: List[int] = Field(title="Class", description="The class of detected item")
labels: List[str] = Field(title="Label", description="The label of detected item")
boxes: List[List[int]] = Field(title="Box", description="The bounding box of detected item")
images: List[str] = Field(title="Image", description="The base64 encoded images of detected faces")
scores: List[float] = Field(title="Scores", description="The scores of the detected faces")
classes: list[int] = Field(title="Class", description="The class of detected item")
labels: list[str] = Field(title="Label", description="The label of detected item")
boxes: list[list[int]] = Field(title="Box", description="The bounding box of detected item")
images: list[str] = Field(title="Image", description="The base64 encoded images of detected faces")
scores: list[float] = Field(title="Scores", description="The scores of the detected faces")
class ResMask(BaseModel):
mask: str = Field(default='', title="Image", description="The processed image in base64 format")
@@ -47,13 +46,13 @@ class ItemPreprocess(BaseModel):
params: dict = Field(title="Params")
class ItemMask(BaseModel):
models: List[str] = Field(title="Models")
colormaps: List[str] = Field(title="Color maps")
models: list[str] = Field(title="Models")
colormaps: list[str] = Field(title="Color maps")
params: dict = Field(title="Params")
types: List[str] = Field(title="Types")
types: list[str] = Field(title="Types")
class APIProcess():
class APIProcess:
def __init__(self, queue_lock: Lock):
self.queue_lock = queue_lock
modules/api/script.py
+1 -2
View File
@@ -1,4 +1,3 @@
from typing import Optional
from fastapi.exceptions import HTTPException
import gradio as gr
from modules.api import models
@@ -36,7 +35,7 @@ def get_scripts_list():
return models.ResScripts(txt2img = t2ilist, img2img = i2ilist, control = control)
def get_script_info(script_name: Optional[str] = None):
def get_script_info(script_name: str | None = None):
res = []
for script_list in [scripts_manager.scripts_txt2img.scripts, scripts_manager.scripts_img2img.scripts, scripts_manager.scripts_control.scripts]:
for script in script_list:
modules/api/xyz_grid.py
+2 -3
View File
@@ -1,7 +1,6 @@
from typing import List
def xyz_grid_enum(option: str = "") -> List[dict]:
def xyz_grid_enum(option: str = "") -> list[dict]:
from scripts.xyz import xyz_grid_classes # pylint: disable=no-name-in-module
options = []
for x in xyz_grid_classes.axis_options:
@@ -23,4 +22,4 @@ def xyz_grid_enum(option: str = "") -> List[dict]:
def register_api():
from modules.shared import api as api_instance
api_instance.add_api_route("/sdapi/v1/xyz-grid", xyz_grid_enum, methods=["GET"], response_model=List[dict])
api_instance.add_api_route("/sdapi/v1/xyz-grid", xyz_grid_enum, methods=["GET"], response_model=list[dict])