Enables an automated MLOps pipeline for Stable Diffusion model fine-tuning using multiple Google Cloud services, including Vertex AI, Cloud Storage, Cloud Build, PubSub, Firestore, Cloud Run, and Cloud Functions.
Handles image storage for training data, maintains predefined bucket paths for uploads, and stores compiled pipeline artifacts for Stable Diffusion fine-tuning jobs.
Used to create notebooks that outline pipeline workflows and components for Stable Diffusion model fine-tuning on Vertex AI.
Used in cloud functions to subscribe to topics and trigger Vertex AI pipeline jobs for Stable Diffusion model fine-tuning.
Provides the frontend portal interface for uploading training images, deployed as a Cloud Run service to interact with the backend processing pipeline.
Handles compiled pipeline definitions that are stored in Google Cloud Storage and used to trigger Vertex AI training jobs.
sd-for-designers
A fully automated workflow for triggering, running & managing fine tuning, training & deploying custom stable diffusion models using Vertex AI
Description
Sd-aa-S is a full automated MLOps pipeline for triggering, managing & tracking Stable diffusion finetuning jobs on GCP using GCP components such as Google Cloud Storage, Cloud Build, Cloud PubSub, Firestore, Cloud Run, Cloud Functions and Vertex AI. It aims to simplify the ML workflows for tuning Stable diffusion using different techniques, starting with Dreambooth. Support for Lora, ControlNet etc. coming soon. The project is targeted ML/Data Engineers, Data Scientists & anybody else interested in or on the road towards building a platform for finetuning stable diffusion at scale.
Three Parts
1. The App part
2. The Vertex AI part
3. The Plumbing part
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
kjlahsdjkashd
Related MCP Servers
- JavaScript
- TypeScriptMIT License
- RustMIT License
- PythonMIT License