LTM-2-Mini is our first model with a 100M token context window

Is there a research paper or white paper associated with it to know more about the architecture? Is it like a RAG technique where 100M token is similar to storing data (token) externally and fetching relevant pieces? What is the price associated with it? Please share more details. Thank you.

URL should probably be: https://magic.dev/blog/100m-token-context-windows instead.