Blog
Notes on adapters, fine-tuning, and running models on your own hardware.
API Design for Adapter Catalogs
A practical API shape for listing adapters, fetching versions, creating jobs, and downloading artifacts in a small registry.
2025-07-17Understanding Fine-Tune Queue States
How to represent fine-tune job states clearly so users know whether a run is queued, active, blocked, failed, or complete.
2025-07-11Building a Lightweight Inference Playground Without Heavy Infrastructure
How to design a usable model playground with prompt input, progress states, and sample outputs without a large backend stack.
2025-07-03Prompt Sets for Image Regression Testing
How to build fixed prompt sets for image adapters so version comparisons stay meaningful and repeatable.
2025-06-19Versioning and Sharing Adapters
How to release adapters predictably with hashes, tags, compatibility notes, and changelogs.
2025-06-18Using Adapters in ComfyUI
How LoRA adapters fit into ComfyUI workflows and what to check before loading them.
2025-06-05Memory Requirements for LoRA Training
How to estimate GPU memory for LoRA jobs and avoid the usual out-of-memory failures.
2025-06-02
Getting Started with LoRA Adapters
What LoRA adapters are, how they fit into a model registry, and how to use them without wasting time or disk space.
2025-05-28Image vs Text Adapters
What changes when a registry serves both image and text adapters under the same product surface.
2025-05-20Public vs Private Adapters in a Team Registry
How to separate public and private adapters in one registry without confusing visibility, permissions, or release flow.
2025-05-14