$lightweight --version 2.0

One API key.
Every model.

Every frontier model. One endpoint. Zero config.

api.lightweight.one
$features.list

Built for Production

Everything you need to ship AI applications with confidence.

Smart Routing

Automatic model selection and context overflow handling. Requests over 160K tokens auto-route to extended context backends.

Production-Grade

Pool rotation, auto-retry with exponential backoff, circuit breaker, and rate limiting. Built on edge infrastructure.

30+ Models

Claude, GPT, Gemini, Grok, and more. Chat completions, Responses API, and Embeddings — all through a single endpoint.

OpenAI-Compatible

Drop-in replacement for the OpenAI API. Works with any SDK, CLI tool, or IDE extension out of the box.

$demo.run --interactive

Drop-in Compatible

Use any OpenAI SDK. Just change the base URL.

terminal
curl https://api.lightweight.one/v1/chat/completions \
  -H "Authorization: Bearer lw_sk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'