RootInference

Unified API

One OpenAI-compatible endpoint for 500+ models from 60+ providers

Change the base URL, keep your SDK. RootInference exposes a single OpenAI-compatible API that speaks to 500+ models from 60+ providers — streaming, tool calling, structured outputs and vision included. No provider-specific clients, no per-vendor billing, no rewrites.

Get an API keyFrom <1 credit per request
Tracecurl https://api.rootinference.com/v1/chat/completions
  1. Request received for anthropic/claude-sonnet-4-5, streaming on

  2. Resolved via unified schema — no client changes required

  3. Tool call emitted: get_weather, city Berlin

  4. First token in 212 ms at 96 tok/s

  5. Usage metered: 1,412 in / 388 out, priced at 0.9 credits

  6. Same request shape works on 500+ models

Capabilities

What it does

Integrate once and reach every frontier and open model — switching providers becomes a string change.

  • Drop-in compatibility with existing OpenAI SDKs — Python, Node, Go and more
  • 500+ models across OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek and 60+ providers
  • Streaming responses with identical event semantics across every model
  • Tool calling and function calling normalised across providers
  • Structured outputs and JSON mode on models that support them
  • Vision and multimodal inputs through the same chat endpoint
  • One API key, one bill, one usage ledger across every model you use

See Unified API on your own traffic

Point your existing OpenAI SDK at our base URL and try it in minutes — or book a demo to plan a production rollout.