Thread

Does this mean you’re more open to cashu?

What does this have to do with cashu

you didn't read the RIPs, neither did I 😅

Does this mean you're more open towards trimming overgrown hooves to horses?

Great! Take a look at local models, I just added multiple locally running models to nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyv9kh2uewd9hj7qghwaehxw309aex2mrp0yh8qunfd4skctnwv46z7qg6waehxw309aexjcn09ejh2tnwdaehgunfvyhxzurs9uqzp5daxvenwv7ucsglpm5f8vuts530cr0zylllgkwejpzvak0x2kqmehusjv Needs more polishing, was up late last night getting it done. Can do translations, sentiment, speech to text, text to speech, with full privacy, all running locally. These models are just 60-150MB, works good enough. Text generation and chat can be done too but require 2-3GB models.

decentralized AI sounds challenging

If you do this I recommend you to use MDK for encrypting the conversation. That way at least there's **some** privacy. https://github.com/parres-hq/mdk

thats so complex though... sure that's nice for very secure DMs, but it doesn't make sense for everything

Honestly it's not that hard to use, and if you want LLM conversations over nostr they better be encrypted at least on my opinion

of course, we are actively discussing encryption options. i designed pns nostr:nevent1qqsyrusel2t8lu9a3hmaxf9w8dplx462eav24l77c6n74g59c6m4t2cppemhxue69uhkummn9ekx7mp0qyghwumn8ghj7mn0wd68ytnhd9hx2tcpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhsz9mhwden5te0wfjkccte9ehx7um5wghxyctwvshsrqpxg0 specifically for this private self backup usecase. can just use giftwraps for handing off convos to specific users

Haven't seen that before, thanks for sharing

🥳🥳🥳