AI Blocks › Build and ship software
Picture chat assistant builder
Give it images and questions. Get a chat that describes pictures and answers questions about them.
Picture chat assistant builder is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · MIT licence · Made by Orchestra Research · Level: advanced
Made by Orchestra Research. See the original on GitHub; we copied it as it is, with its licence.
What it does
A chat assistant that looks at pictures and talks about them. Ask what is in a photo, what a chart says or what a document image contains, and keep asking follow-ups. The skill sets up LLaVA, an open-source model that pairs vision with a language model. You clone the project, install it, load a model in a chosen size and feed it images with your question. It supports multi-turn chats, captions, visual question answering and following instructions about an image. It is a good fit when you want to run it yourself and pay no per-call fee. It needs Python and a GPU.
You give it
- Images
- Questions to ask
- A computer with a GPU
- Model size choice (optional)
You get
- A running picture chat setup
- Image descriptions
- Answers to questions about each image
- Multi-turn conversations
Use it when
- You want a bot that answers questions about product photos.
- You need captions for a pile of images.
- You want to ask follow-up questions about a screenshot.
- You prefer running a model yourself over a paid service.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run Python code on a computer with a GPU.