AI Infrastructure
76 infrastructure entities tracked · APIs and frameworks · updated daily
43 AI APIs and 33 frameworks tracked across 46 providers. Infrastructure entities carry the highest relation density in the ecosystem graph.
↳ Explore AI infrastructure →
Programmatic interfaces exposing AI capabilities and data. LLM inference, embedding, speech, data APIs.
Explore AI APIs →SDKs, agent frameworks, and developer libraries. LangChain, official SDKs, evaluation frameworks.
Explore AI Frameworks →Infrastructure at the core of the graph
integrates_with relations — the most active relation type connecting software to the infrastructure beneath it
entity_relations · 2026-04-09
What is AI Infrastructure?
AI infrastructure is the layer that models and software are built on. sourc.dev tracks 76 infrastructure entities — from inference APIs to agent frameworks.
What qualifies as an AI API?
A programmatic interface exposing AI or data capabilities via HTTP. No UI required. Includes LLM inference APIs, embedding APIs, speech APIs, and data APIs. sourc.dev tracks 43 AI APIs.
What is an AI Framework?
A library or SDK for building AI applications. Includes LLM integration frameworks like LangChain, agent frameworks, official model SDKs, and evaluation libraries. sourc.dev tracks 33 frameworks.
How do APIs differ from SaaS?
An API exposes capabilities programmatically — no UI, accessed via HTTP. A SaaS platform has a UI, user accounts, and subscription pricing. Products that expose both are tracked in the branch reflecting primary value.
Data verified from primary sources. Methodology →