Blog
Notes from the quarry
What we found digging through open-source alternatives, what changed, and how we check.
RSS feed
Be more ambitious with AI coding: bigger outcomes, smaller proofs
A practical RepoQuarry perspective on ambitious AI development: define the outcome, choose open source tools for the job and build a small experiment with evidence you can check.
Read the post →-
ArtCraft vs Adobe: what is worth testing before you switch?
ArtCraft’s creative apps have real installers and enthusiastic users, but early builds still need testing. We compare official claims, community reports and the checks that matter before replacing an Adobe task.
-
October open source release watch: Ollama and Immich
Two October updates worth understanding: decision models in Ollama and people-sharing fixes in Immich. What shipped, what it means and what to test.
-
How to prompt AI to find better open source alternatives
A reusable prompt for finding software alternatives with evidence, practical constraints and a clear trial plan instead of another vague top ten list.
-
Three open source repos worth testing before your next renewal
Ollama, whisper.cpp and Penpot solve three different jobs. Here is how to decide whether any of them deserves a place in your workflow.
-
Comfy Agent: what the cloud launch means for local AI creators
Comfy Agent brings natural-language help to visual workflows. Here is what its October launch actually offered, what was planned and how to evaluate it.
-
SoL-Refiner: why faster AI video needs a closer look
A look back at late September video-refinement research, and the checks that turn an impressive speed claim into a useful creator decision.
-
Ming Image Design: look beyond the impressive AI poster demo
A September image-model spotlight with a practical question: can the output survive real text, brand colours and the next edit?