LazySusan has an API and a Python SDK that put chat, image and video models behind one key and one prepaid balance. Switching models means changing one string.

from lazysusan import LazySusan client = LazySusan() # reads LAZYSUSAN_API_KEY  reply = client.chat.create( model="anthropic/claude-sonnet-5-5", messages=[{"role": "user", "content": "Summarise this in two lines: ..."}], ) print(reply.text) print(reply.usage.cost_credits)

What it covers, per the docs:

  • Streaming, tool calling, JSON output and image input
  • Async video generation
  • Token usage and credit cost on every response
  • Per-key monthly budgets, model allowlists and expiry dates
  • Metadata tags so you can filter usage by project or feature

The Python SDK has zero dependencies, is typed, and runs on Python 3.9+.

Install: pip install lazysusan

Docs: https://www.lazysusan.ai/developers/docs

Disclosure: I built this.