Instant Search is the first accurate local AI file search that uses the latest embedding models from Perplexity to find the exact file (and page) with the content you’re looking for with a 95% hit rate.
Search never worked properly, especially with local files. Now, you can find the right image, document, or Obsidian file in seconds. Looking for your driver’s license? Just type “driver’s license”. Looking for a particular note? Just explain the context.
Why not Finder/Spotlight?
Spotlight is just a keyword search. It doesn’t understand file content. Instant finds the exact file you’re describing in under a second.
Why not RAG?
Traditional RAG models give you a single embedding vector per entire body of text. That means either you have a vector that’s not detailed enough to give you an accurate semantic representation, or it’s too precise and doesn’t contain enough surrounding context. Instant circumvents these problems by relying on a bidirectional late-interaction embedding model that works flawlessly.
How much space does indexing need? An average library of content needs less than 1GB of disk space.
Check it out, share with friends, and let me know your feedback! It's free.