Framework

Trusted Provider as Security Layer

A model is text in, text out. Run the weights on a trusted provider and the request cannot leave that server.

How it works

  1. 1Run it on a provider you trust and the Chinese government does not get your data, and neither does anyone else.
  2. 2Censorship is real on hot-button topics, so do not let these models write your policy. For generating HTML for your video game, it is a non-issue.
  3. 3Bifurcate: route sensitive work to a US frontier model you trust, route everything else to cheaper open models, and run them in tandem.

Hear it on the podcast

40,000 Models, One API Key, and a $25M Bet on Open SourceListen to the episode

More from this episode

Get the next framework first

One email a week with the frameworks, tools, and founder stories from the podcast. Free.

More frameworks for AI & Automation