Thanks for putting together this repo! The RLM concept is awesome, and your code made it super easy to figure out how it all works under the hood.
I liked the idea so much that I decided to bring those token-saving mechanics to the pi ecosystem. I just built and published a plugin for it:
🔗 GitHub: openzebra/rlm.pi
🧩 Pi Package: @hicaru/pi-rlm
📦 npm: @hicaru/pi-rlm
It uses the rlm_query logic just like in the docs, and it actually saves a ton of tokens in real-world usage. Works like a charm!
Just wanted to drop this here to say thanks for the inspiration. Also, if you ever decide to add a "Community Implementations" or "Ecosystem" section to your README, I'd love to submit a PR to add it there.
Cheers!
Thanks for putting together this repo! The RLM concept is awesome, and your code made it super easy to figure out how it all works under the hood.
I liked the idea so much that I decided to bring those token-saving mechanics to the
piecosystem. I just built and published a plugin for it:🔗 GitHub: openzebra/rlm.pi
🧩 Pi Package: @hicaru/pi-rlm
📦 npm: @hicaru/pi-rlm
It uses the
rlm_querylogic just like in the docs, and it actually saves a ton of tokens in real-world usage. Works like a charm!Just wanted to drop this here to say thanks for the inspiration. Also, if you ever decide to add a "Community Implementations" or "Ecosystem" section to your README, I'd love to submit a PR to add it there.
Cheers!