Content
@
https://warpcast.com/~/channel/pinata
0 reply
0 recast
0 reaction
Justin Hunter
@polluterofminds
Should I publish it on the Pinata blog? What's happening here? There's a GPU-powered machine running somewhere in the smoky mountains and it has a crypto wallet. That machine has a locally running llama 3.2 instance. That wallet uses x402 protocol to request payment for people making inference requests to the model. Back home in the flat expanse of Dallas, I have a script using the OpenAI SDK, modified to use the x402 payments wrapper. It makes a payment to the machine in the mountains via crypto before the machine spits out its AI generated response. Full OpenAI spec compatible.
0 reply
4 recasts
6 reactions
