Open-Source Frameworks & Local Tooling
This is the plumbing layer of the AI stack: open-source libraries and infrastructure for running, fine-tuning, and retrieving against models locally or in your own pipeline. Hugging Face anchors the category as the de facto model/dataset hub; PyTorch, vLLM, and Ollama cover training and local inference; vector databases and retrieval libraries (the RAG stack) round it out. What unites these companies/projects is that builders integrate them directly into code rather than consuming them as a black-box API — so version changes, breaking API updates, and new feature releases here have immediate, hands-on consequences for anyone building on top. GROUNDING tracks releases, breaking changes, and the retrieval/fine-tuning techniques (chunking, hybrid search, reranking) this tooling exists to support.
At a glance
- 7 tracked companies (5 hand-curated)
- Most recently updated: ONNX Runtime (2026-06-22)
Most actively covered
- Hugging Face ★
- LangChain ★
- Qdrant ★
- Pinecone ★
- Weaviate ★
- Framework
- ONNX Runtime
FAQ
What is the Open-Source Frameworks & Local Tooling category?
Publishes open libraries, runtimes, or infrastructure for building or running models.
How many companies are in this category?
GROUNDING currently tracks 7 companies in Open-Source Frameworks & Local Tooling, 5 of which have a hand-curated profile.
How current is this hub?
The most recently updated entry is ONNX Runtime, last mentioned 2026-06-22; the radar refreshes hourly.