**My first Jev result!** Recreated the example "Line-by-line search" ( ) to process real Terms of Service (ToS) docs,

**My first Jev result!** Recreated the example "Line-by-line search" (https://docs.typesafe.ai/cookbooks/semantic_find) to process real Terms of Service (ToS) docs, and identifying the specific lines of the ToS that were related to user or vendor ownership of input/output content (AI Vendor ToS'), and then compared that to the same function being run by an LLM (DeepSeek V4 Flash)... the results were immediate **23x faster with Jev!** Also the LLM had a false positive (identifying a line about software intellectual property) which **Jev did not! ** However the token count seems high, really high, on the Jev side, need to look into if this is an error, or if the Line-by-Line search is just that inefficient, if it is right the token cost might actually be higher with Jev even with the super low input-only pricing. I'll share more after some debugging and my full POC is running. 😁