This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Tony Sullivan
tonysull.co
did:plc:3g7h3j2hxd3tgay76gwp6bkr
During inference, after each token the entire context including the latest token is sent back through to guess the next one
In "thinking" mode one LLM is doing this whole process itself to prompt another LLM (2/3)
2026-02-24T03:30:40.494Z