Hey everybody, I moved the Wednesday call from 1pm EST to 7pm EST. I got a new job that takes up my day time hours. For the call today I'm going to cover What is AI Inference?

This will give us the ability to communicate much more clearly, talk about what cool projects we're working on, have community sharing events, and be in touch more. Discord is free. Join below. https://discord.gg/CnRBRg2pm

I am looking to pay someone to spend about 2 hours or so with me remotely to to help me not Setup Hermes which I’ve just done but to maximise it and get it working better, it’s hosted on hosttinger, set up a Mission Control I have tried one and a second brain, is this the right forum to ask

Check out pxpipe, a new open-source tool designed to slash Anthropic API costs. Since Anthropic bills based on pixel count rather than word count for image inputs, this repo converts Claude prompts into optimized images before sending them. By leveraging this technique, the creator managed to reduce their total usage by at least 10x. It is an ingenious way to bypass the standard token-heavy billing and make high-volume prompt engineering significantly more affordable. I feel like this will become more useful when they move the model out of subscription. Be warned this is a lossy technique according to their own documentation. But only way to find out if it truly works for your use case is to try it out :)

this is crazy! this model is notably better than anything i have ever used even on /high effort is enough what are you guys using fable 5 for? you now have it through the 12th