← Back to context

Comment by frog437

4 months ago

depending on how large your codebase is, hopefully not. At this point use something like the IX plugin to ingest codebase and track context, rather than from the LLM itself.

2 comments

frog437

Reply
frog437  4 months ago

This is crazy..

tokensSaved = naiveTokens - actualTokens

  - naiveTokens = 19.4M — what ix estimates it would have cost to answer your queries without graph intelligence (i.e., dumping full files/directories into context)                                    
  - actualTokens = 4.7M — what ix's targeted, graph-aware responses actually used
  - tokensSaved = 14.7M — the difference

andoando  4 months ago

I mean whatever part of the code that is read by the AI has to be in the content window at some point or another nSprewd throughout your sessions Id think even with a huge codebase, 90% of it is going to be there