{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui","slug":"divamgupta--diffusionbee-stable-diffusion-ui","name":"diffusionbee-stable-diffusion-ui","type":"model","url":"https://diffusionbee.com","page_url":"https://unfragile.ai/divamgupta--diffusionbee-stable-diffusion-ui","categories":["image-generation"],"tags":["electron-app","macos","stable-diffusion"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_0","uri":"capability://image.visual.local.text.to.image.generation.with.stable.diffusion","name":"local-text-to-image-generation-with-stable-diffusion","description":"Generates images from natural language text prompts by running the Stable Diffusion model entirely on the user's local machine. The backend loads pre-trained PyTorch checkpoints, tokenizes text input through a CLIP text encoder, and iteratively denoises latent representations over configurable diffusion steps to produce final images. All computation happens on-device without cloud API calls, ensuring complete data privacy and offline capability.","intents":["I want to generate images from text descriptions without sending data to external servers","I need to run image generation locally on my Mac without installing complex dependencies","I want to iterate quickly on image generation with full control over model parameters"],"best_for":["macOS users (Intel and Apple Silicon) seeking privacy-first image generation","creative professionals wanting offline, deterministic image synthesis","developers prototyping generative AI features without cloud costs"],"limitations":["macOS-only (no Windows or Linux support in the packaged application)","Generation speed depends on local hardware; M1/M2 Macs significantly faster than Intel","Requires 8-16GB RAM minimum for smooth operation with larger models","Model download sizes range from 4-7GB per variant, consuming significant disk space"],"requires":["macOS 10.13 or later","4GB+ available disk space per model","8GB+ system RAM (16GB+ recommended for batch operations)"],"input_types":["text (natural language prompts)","numeric parameters (guidance scale, steps, seed)"],"output_types":["PNG images (configurable resolution, typically 512x512 or 768x768)"],"categories":["image-visual","local-processing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_1","uri":"capability://image.visual.image.to.image.conditional.generation","name":"image-to-image-conditional-generation","description":"Transforms existing images by encoding them into the latent space and applying conditional diffusion guided by a new text prompt. The system loads the input image, passes it through the VAE encoder to obtain latent representations, then runs the diffusion process starting from a noisy version of these latents (controlled by a strength parameter) while conditioning on the new prompt. This enables style transfer, content modification, and creative reinterpretation without full regeneration.","intents":["I want to modify an existing image based on a text description while preserving its overall composition","I need to apply artistic style changes to photos without starting from scratch","I want to iterate on image variations by tweaking prompts while maintaining structural similarity"],"best_for":["digital artists and designers iterating on visual concepts","content creators needing rapid image variations","users wanting to repurpose existing photos with AI enhancement"],"limitations":["Strength parameter (0-1) controls noise injection; values too low preserve original too much, too high lose coherence with input","Input image resolution must match model training resolution (typically 512x512); upscaling/downscaling may degrade results","Requires the same model variant as text-to-image; cannot mix SD 1.5 and SDXL models in a single operation"],"requires":["macOS 10.13 or later","Stable Diffusion model loaded in memory","Input image in PNG, JPG, or WebP format"],"input_types":["image (PNG, JPG, WebP)","text prompt (natural language)","numeric parameters (strength 0-1, guidance scale, steps)"],"output_types":["PNG image (same resolution as input)"],"categories":["image-visual","local-processing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_10","uri":"capability://memory.knowledge.image.gallery.and.generation.history.management","name":"image-gallery-and-generation-history-management","description":"Maintains a local gallery of generated images with metadata (prompt, parameters, timestamp, model used) and enables browsing, searching, and organizing results. The system stores images in a local directory structure, indexes metadata in a JSON database, and provides UI components for filtering by date, model, or prompt keywords. Users can favorite images, delete batches, export results, and view detailed generation parameters for reproducibility.","intents":["I want to keep track of all images I've generated and the parameters used to create them","I need to search my generation history to find a specific image or prompt","I want to export batches of images for use in other projects"],"best_for":["artists maintaining a portfolio of generated work","researchers tracking experimental results and parameters","content creators organizing assets for projects"],"limitations":["Gallery performance degrades with >10,000 images; no pagination or lazy loading in current implementation","Metadata is stored in JSON; no full-text search or advanced query capabilities","Deleting images from the gallery doesn't free disk space (requires manual cleanup)","No cloud sync; gallery is local-only and not backed up automatically"],"requires":["macOS 10.13 or later","Local disk space for image storage (1-2MB per image)","Electron 12+ (included in application)"],"input_types":["image metadata (JSON with prompt, parameters, timestamp)","search queries (text strings, date ranges)"],"output_types":["gallery view (HTML/Vue.js component)","image metadata (JSON)","exported images (ZIP file or directory)"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_11","uri":"capability://automation.workflow.one.click.installer.with.dependency.bundling","name":"one-click-installer-with-dependency-bundling","description":"Provides a single-click macOS installer that bundles all dependencies (Python runtime, PyTorch, model files) into a self-contained application package. The installer uses Electron's native packaging tools to create a .dmg file that users can mount and drag into Applications. On first launch, the application downloads required models and configures the Python environment automatically. No manual dependency installation, environment variables, or terminal commands are required.","intents":["I want to install Stable Diffusion on my Mac without using the terminal or installing Python","I need a simple installer that handles all dependencies automatically","I want to uninstall the application cleanly without leaving behind configuration files"],"best_for":["non-technical users unfamiliar with Python or command-line tools","macOS users wanting a native application experience","organizations deploying Stable Diffusion to multiple machines without IT support"],"limitations":["Installer size is large (~2-3GB) due to bundled Python and dependencies","First-run setup requires internet connection for model downloads (can take 10-30 minutes)","No support for custom Python environments or virtual environments","Updates require re-downloading the entire application; no incremental updates"],"requires":["macOS 10.13 or later","Intel or Apple Silicon processor","4GB+ free disk space for application","20GB+ free disk space for models","Internet connection for initial setup"],"input_types":["installer file (.dmg)","user interactions (drag-to-install, first-run setup)"],"output_types":["installed application in /Applications directory","configuration files in ~/Library/Application Support/DiffusionBee/","downloaded models in ~/Library/Application Support/DiffusionBee/models/"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_12","uri":"capability://automation.workflow.apple.silicon.metal.acceleration.for.inference","name":"apple-silicon-metal-acceleration-for-inference","description":"Optimizes image generation performance on Apple Silicon (M1/M2/M3) Macs by leveraging Metal GPU acceleration for PyTorch operations. The system detects the processor type at runtime, configures PyTorch to use Metal Performance Shaders (MPS) backend instead of CPU, and offloads matrix multiplications, convolutions, and attention operations to the GPU. This provides 3-5x speedup compared to CPU-only inference while maintaining compatibility with Intel Macs.","intents":["I want faster image generation on my M1/M2 Mac without additional hardware","I need to understand if my Mac will run Stable Diffusion efficiently","I want to optimize generation speed for real-time iteration"],"best_for":["Apple Silicon Mac users (M1, M2, M3 and later) seeking optimal performance","creative professionals needing fast iteration cycles","users with limited VRAM wanting to maximize throughput"],"limitations":["Metal acceleration only available on Apple Silicon; Intel Macs fall back to CPU (significantly slower)","Some operations (attention mechanisms) may have bugs in Metal backend; fallback to CPU adds latency","Memory bandwidth is shared between GPU and system RAM; large models may still be slow","Metal support in PyTorch is still evolving; newer PyTorch versions may have compatibility issues"],"requires":["macOS 12.0 or later (Metal support)","Apple Silicon processor (M1, M2, M3, or later)","PyTorch 1.12+ with Metal support"],"input_types":["model checkpoint (PyTorch format)","generation parameters (prompt, steps, guidance scale)"],"output_types":["PNG image","performance metrics (generation time, VRAM usage)"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_2","uri":"capability://image.visual.inpainting.selective.image.region.replacement","name":"inpainting-selective-image-region-replacement","description":"Enables selective replacement of masked regions within an image while preserving unmasked areas. Users draw or upload a mask indicating which pixels to regenerate, and the system encodes both the original image and mask into latent space, then runs diffusion only on the masked regions conditioned by the text prompt. The VAE decoder reconstructs the final image with seamless blending between modified and original regions, using specialized inpainting model variants trained to handle mask boundaries.","intents":["I want to remove or replace specific objects in an image without affecting the rest","I need to fix or enhance particular areas of a photo (e.g., remove blemishes, change clothing)","I want to add new elements to specific regions while keeping the background intact"],"best_for":["photo editors and retouchers needing AI-assisted object removal","content creators fixing unwanted elements in images","designers compositing multiple elements into a single image"],"limitations":["Requires inpainting-specific model variants (not compatible with standard text-to-image models)","Mask quality directly affects output quality; soft/feathered edges produce better blending than hard edges","Large masked regions may show visible seams or inconsistencies if the prompt is vague","Mask must be binary or grayscale; color masks are not supported"],"requires":["macOS 10.13 or later","Inpainting-specific Stable Diffusion model variant loaded","Input image and corresponding mask image (same dimensions)"],"input_types":["image (PNG, JPG, WebP) - the base image","mask image (grayscale PNG) - white=regenerate, black=preserve","text prompt (natural language)","numeric parameters (guidance scale, steps)"],"output_types":["PNG image (same resolution as input)"],"categories":["image-visual","local-processing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_3","uri":"capability://image.visual.outpainting.image.extension.beyond.boundaries","name":"outpainting-image-extension-beyond-boundaries","description":"Extends images beyond their original boundaries by padding the canvas and using inpainting to generate new content in the expanded regions. The system resizes the original image to fit within a larger canvas, creates a mask for the new border areas, and runs the inpainting pipeline to synthesize contextually appropriate content that seamlessly blends with the original image edges. This enables creative composition expansion and context generation without cropping.","intents":["I want to extend an image in specific directions (up, down, left, right) to change composition","I need to add more context or background to an existing image","I want to create wider aspect ratios from square images without distortion"],"best_for":["photographers and designers needing to adjust image composition post-capture","content creators adapting images for different aspect ratios (social media, prints)","artists expanding scenes or adding environmental context"],"limitations":["Quality degrades with very large extensions (>50% of original dimensions); small incremental extensions work best","Requires inpainting model variant; cannot use standard text-to-image models","Generated content may not perfectly match original image style if the prompt is too generic","Canvas expansion is limited by available VRAM; very large canvases may cause out-of-memory errors"],"requires":["macOS 10.13 or later","Inpainting-specific Stable Diffusion model variant","Input image in PNG, JPG, or WebP format"],"input_types":["image (PNG, JPG, WebP)","numeric parameters (extension amount per direction, guidance scale, steps)","text prompt (optional; if not provided, uses image analysis to infer context)"],"output_types":["PNG image (larger resolution than input)"],"categories":["image-visual","local-processing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_4","uri":"capability://image.visual.controlnet.conditional.generation.with.structural.guidance","name":"controlnet-conditional-generation-with-structural-guidance","description":"Enables image generation guided by structural conditions (edge maps, depth maps, pose skeletons, semantic segmentation) through ControlNet modules that inject spatial constraints into the diffusion process. The system loads a ControlNet model corresponding to the desired control type, encodes the control image into a conditioning signal, and injects this signal into the UNet at multiple scales during denoising. This allows precise control over image composition, layout, and structure while the text prompt guides semantic content.","intents":["I want to generate images that follow a specific composition or layout defined by an edge map or sketch","I need to create images with consistent pose or perspective as a reference image","I want to control depth distribution or semantic layout while letting the model fill in details"],"best_for":["concept artists and storyboarders needing layout control during generation","game developers creating consistent character poses and perspectives","designers prototyping compositions before detailed rendering"],"limitations":["Requires separate ControlNet model downloads (additional 1-2GB per control type)","Control image quality directly impacts output; noisy or ambiguous control signals produce poor results","Only one ControlNet can be active at a time (no multi-control composition in current implementation)","Control types are model-specific; ControlNet trained on SD 1.5 may not work optimally with SDXL"],"requires":["macOS 10.13 or later","Base Stable Diffusion model loaded","ControlNet model variant for desired control type (canny, depth, pose, segmentation, etc.)","Control image in PNG or JPG format"],"input_types":["control image (edge map, depth map, pose skeleton, or segmentation mask)","text prompt (natural language)","numeric parameters (control strength 0-1, guidance scale, steps)"],"output_types":["PNG image (resolution matching control image)"],"categories":["image-visual","local-processing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_5","uri":"capability://memory.knowledge.multi.model.management.and.switching","name":"multi-model-management-and-switching","description":"Manages multiple Stable Diffusion model variants (SD 1.5, SD 2.0, SDXL, inpainting, LoRA) with dynamic loading and unloading to optimize memory usage. The system maintains a model registry, downloads models from Hugging Face, stores them locally, and loads/unloads them from VRAM on-demand based on user selection. The bridge layer communicates model state changes to the frontend with progress indicators, and the backend handles model compatibility checks to prevent incompatible model/task combinations.","intents":["I want to switch between different Stable Diffusion models without restarting the application","I need to use specialized models (inpainting, SDXL) for different tasks without managing files manually","I want to download and manage multiple models while keeping my disk usage under control"],"best_for":["power users experimenting with multiple model variants and quality tiers","professionals needing different models for different tasks (text-to-image vs inpainting)","users with limited VRAM wanting to swap models rather than run multiple instances"],"limitations":["Model switching requires full unload/reload cycle (~5-30 seconds depending on model size and hardware)","Each model variant requires separate disk storage (4-7GB per model); managing 3+ models requires 20GB+ free space","No built-in model quantization; full-precision models consume maximum VRAM","LoRA models must be manually downloaded and placed in the correct directory; no integrated LoRA marketplace"],"requires":["macOS 10.13 or later","Internet connection for initial model downloads","20GB+ free disk space for multiple models","8GB+ VRAM (16GB+ recommended for smooth multi-model switching)"],"input_types":["model selection (dropdown or file browser)","model download URLs or local file paths"],"output_types":["model state indicators (loading, ready, error)","progress messages during download/load"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_6","uri":"capability://safety.moderation.secure.pytorch.checkpoint.loading.without.pickle","name":"secure-pytorch-checkpoint-loading-without-pickle","description":"Loads PyTorch model checkpoints safely without using pickle deserialization, which can execute arbitrary code during unpickling. The system implements a custom checkpoint loader that parses the model architecture separately from weights, validates the structure against known safe schemas, and loads weights using safe serialization formats (SafeTensors or manual tensor reconstruction). This prevents code injection attacks while maintaining compatibility with standard PyTorch checkpoint formats.","intents":["I want to load community-provided models without worrying about malicious code execution","I need to ensure that model downloads from untrusted sources don't compromise my system","I want to audit what code is being executed when loading models"],"best_for":["security-conscious users downloading models from community sources","organizations with strict security policies requiring code review before model loading","developers building model distribution systems that need to prevent supply-chain attacks"],"limitations":["Custom checkpoint loading adds ~2-5 second overhead per model load compared to standard pickle","Only supports model architectures that have been explicitly whitelisted (SD 1.x, SD 2.x, SDXL)","Custom architectures or experimental models may fail to load if their structure isn't recognized","SafeTensors format support requires models to be pre-converted; older pickle-only checkpoints cannot be loaded"],"requires":["macOS 10.13 or later","PyTorch 1.9+ with SafeTensors support","Model checkpoint in SafeTensors format or standard PyTorch format with known architecture"],"input_types":["model checkpoint file (SafeTensors or PyTorch .pt/.pth)","model architecture specification (JSON schema)"],"output_types":["loaded PyTorch model object","validation report (architecture matches schema, no unknown layers)"],"categories":["safety-moderation","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_7","uri":"capability://image.visual.interactive.canvas.image.manipulation.tools","name":"interactive-canvas-image-manipulation-tools","description":"Provides a browser-based canvas interface for drawing masks, selecting regions, and manipulating images before generation. The frontend uses HTML5 Canvas and Vue.js to implement brush tools, selection tools, layer management, and real-time preview. Users can draw masks for inpainting, adjust image parameters (brightness, contrast, saturation), crop/resize, and compose multiple layers. The canvas state is serialized and sent to the backend for processing, with pixel-level precision maintained throughout.","intents":["I want to draw precise masks for inpainting without external image editing software","I need to adjust image parameters and preview changes in real-time before generation","I want to compose multiple images or layers and apply generation to specific regions"],"best_for":["digital artists and designers wanting integrated editing within the generation workflow","users without access to Photoshop or GIMP who need basic image manipulation","creators iterating quickly on image composition without context-switching between apps"],"limitations":["Canvas tools are basic compared to professional image editors (no advanced filters, no non-destructive editing)","Large images (>2048x2048) may cause performance degradation due to JavaScript rendering overhead","No layer blending modes beyond basic opacity; advanced compositing requires external tools","Undo/redo history is limited to ~20 steps to conserve memory"],"requires":["macOS 10.13 or later","Electron 12+ (included in application)","Modern browser engine (Chromium-based, included in Electron)"],"input_types":["mouse/trackpad input (brush strokes, selections)","image files (PNG, JPG, WebP) for import","numeric parameters (brush size, opacity, color)"],"output_types":["canvas state (serialized as JSON + base64 image data)","mask image (grayscale PNG)","edited image (PNG)"],"categories":["image-visual","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_8","uri":"capability://tool.use.integration.electron.ipc.bridge.frontend.backend.communication","name":"electron-ipc-bridge-frontend-backend-communication","description":"Establishes bidirectional communication between the Electron frontend (Vue.js) and Python backend using a message-based protocol with standardized operation codes. The bridge uses Electron's IPC (Inter-Process Communication) to send requests from the frontend and receive responses/progress updates from the backend. Messages include operation codes (mltl, mlpr, mdld, inwk, dnpr, nwim, inrd), payload data, and metadata. This architecture keeps the UI responsive during long-running operations by using asynchronous message passing.","intents":["I want the UI to remain responsive while the backend processes image generation","I need to receive real-time progress updates during model loading and image generation","I want to cancel long-running operations from the frontend without freezing the application"],"best_for":["Electron application developers building desktop AI tools","teams needing to bridge JavaScript frontends with Python backends","developers wanting to keep UI responsive during computationally intensive operations"],"limitations":["IPC message passing adds ~5-10ms latency per message; high-frequency updates (>100/sec) may cause bottlenecks","Large payloads (>100MB images) require chunking or base64 encoding, increasing memory overhead","No built-in message queuing; rapid requests may be dropped if the backend is overloaded","Debugging IPC communication requires Electron DevTools; standard Python debuggers cannot inspect messages"],"requires":["Electron 12+","Python 3.8+","Node.js 14+ (included with Electron)"],"input_types":["JSON messages with operation codes and payloads","binary data (images, model files) encoded as base64 or file paths"],"output_types":["JSON responses with status codes and results","progress messages with percentage/step indicators","error messages with stack traces"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-divamgupta--diffusionbee-stable-diffusion-ui__cap_9","uri":"capability://automation.workflow.batch.image.generation.with.parameter.variation","name":"batch-image-generation-with-parameter-variation","description":"Generates multiple images in sequence with varying parameters (different prompts, seeds, guidance scales, or model variants) without requiring manual re-submission for each image. The system queues generation requests, processes them sequentially on the backend, and streams results back to the frontend as they complete. Users can define parameter grids (e.g., 3 prompts × 4 seeds = 12 images) and the system automatically expands and processes them, with progress tracking for each batch.","intents":["I want to generate multiple image variations with different seeds to find the best result","I need to test how different prompts affect the output without manual resubmission","I want to create a grid of images with varying parameters for comparison"],"best_for":["artists and designers exploring parameter space to find optimal settings","researchers evaluating model behavior across different inputs","content creators generating multiple assets for projects in bulk"],"limitations":["Batch processing is sequential, not parallel; total time = sum of individual generation times","Large batches (>100 images) may consume significant disk space for results storage","No built-in result filtering or ranking; users must manually review all outputs","Batch jobs cannot be paused/resumed; cancellation loses all queued items"],"requires":["macOS 10.13 or later","Sufficient disk space for batch results (typically 1-2MB per image)","Stable Diffusion model loaded in memory"],"input_types":["batch configuration (JSON with parameter grids)","prompt list (array of text strings)","numeric parameters (seed range, guidance scale range, step count)"],"output_types":["PNG images (one per parameter combination)","batch metadata (JSON with parameters used for each image)","progress log (text file with timestamps and status)"],"categories":["automation-workflow","image-visual"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":38,"verified":false,"data_access_risk":"high","permissions":["macOS 10.13 or later","4GB+ available disk space per model","8GB+ system RAM (16GB+ recommended for batch operations)","Stable Diffusion model loaded in memory","Input image in PNG, JPG, or WebP format","Local disk space for image storage (1-2MB per image)","Electron 12+ (included in application)","Intel or Apple Silicon processor","4GB+ free disk space for application","20GB+ free disk space for models"],"failure_modes":["macOS-only (no Windows or Linux support in the packaged application)","Generation speed depends on local hardware; M1/M2 Macs significantly faster than Intel","Requires 8-16GB RAM minimum for smooth operation with larger models","Model download sizes range from 4-7GB per variant, consuming significant disk space","Strength parameter (0-1) controls noise injection; values too low preserve original too much, too high lose coherence with input","Input image resolution must match model training resolution (typically 512x512); upscaling/downscaling may degrade results","Requires the same model variant as text-to-image; cannot mix SD 1.5 and SDXL models in a single operation","Gallery performance degrades with >10,000 images; no pagination or lazy loading in current implementation","Metadata is stored in JSON; no full-text search or advanced query capabilities","Deleting images from the gallery doesn't free disk space (requires manual cleanup)","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.364129021733025,"quality":0.5,"ecosystem":0.48999999999999994,"match_graph":0.25,"freshness":0.52,"weights":{"adoption":0.35,"quality":0.2,"ecosystem":0.1,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.550Z","last_scraped_at":"2026-05-03T13:58:42.318Z","last_commit":"2024-10-30T16:56:14Z"},"community":{"stars":13560,"forks":720,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=divamgupta--diffusionbee-stable-diffusion-ui","compare_url":"https://unfragile.ai/compare?artifact=divamgupta--diffusionbee-stable-diffusion-ui"}},"signature":"t3kF+isAbL2njhwugJPJLMCZbrskf1T9s/8PWBLuvL4GYeAu5PoW3YQeADLiMnzDaqlQ7OiRqrj0bQ5F4ij+Ag==","signedAt":"2026-06-22T02:17:07.363Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/divamgupta--diffusionbee-stable-diffusion-ui","artifact":"https://unfragile.ai/divamgupta--diffusionbee-stable-diffusion-ui","verify":"https://unfragile.ai/api/v1/verify?slug=divamgupta--diffusionbee-stable-diffusion-ui","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}