🚀 ModelDock — the package manager for local AI models

Running LLMs locally is powerful, but the workflow is fragmented.

Every runtime — Ollama, LM Studio, llama.cpp, Jan, GPT4All, vLLM — has its own way of naming, discovering, pulling, and loading models.

You end up memorizing runtime-specific commands and hunting across ollama.com, Hugging Face Hub, and scattered docs just to find and install a model.

🛠️ What we're building

ModelDock is a lightweight, Python-first management layer that sits above these runtimes.

One clean API and CLI:

import modeldock as md

client = md.load("llama3")   # discover → install if missing → verify → load
md.search("coding")          # search across live sources at once
md.info("qwen3")             # metadata, variants, and where it came from

🔗 GitHub: (https://github.com/OpenAgentHQ/modeldock)

🤝 Looking for Contributors

We're looking for open-source contributors interested in:

• 🤖 Local AI / LLM infrastructure
• 🧠 Model runtimes & providers
• 🐍 Python development
• 🛠️ Developer tooling
• 🔍 Model discovery & management

If this sounds interesting, check out the repo, open an issue, or submit a PR.

Contributions, ideas, and feedback are welcome! ❤️