AI/ML API

AI/ML API is an API gateway that exposes more than 1,000 AI models from many providers through a single endpoint and a single API key. The interface is compatible with the OpenAI and Anthropic APIs, so an existing integration switches over by changing the base URL to api.aimlapi.com/v1, and standard schemas for tool calling, structured outputs and image and PDF input carry across. Model coverage spans chat, code, image, video, audio, voice, search, 3D generation, OCR and safety moderation. It is built for developers, startups and enterprises that want one integration and one bill rather than accounts with every model vendor. An AI Playground lets a team test models before writing code, MCP support connects the gateway to Claude, Cursor and Claude Code, and the documentation covers the OpenAI SDK, Python, Node.js, a command line client, asynchronous calls and integrations with LangChain, LiteLLM, n8n and others. The Enterprise tier adds dedicated servers, unlimited requests and tokens per minute, a 99.9 percent uptime commitment and round the clock support. Pricing is pay as you go, charged per model per million tokens, with a $20 minimum top up on the standard plan and a free playground tier for evaluation. The pricing page lists each model's input and output rate; for example, Claude 4.5 Sonnet is listed at $4.1262 per million input tokens and $20.631 per million output tokens. Volume discounts apply at higher usage, and Enterprise is custom priced. The vendor names Adobe, NVIDIA, ByteDance and Alibaba among its partners, and the roster of models on the pricing page changes often, so a team standardizing on the gateway should check that its preferred models remain listed.