Twas only a matter of time

I was paying per query. Didn’t realize I could use my subscription. Oh well, it didn’t last anyway. 🥲

they don't have the GPUs to sustain the usage

Been using OAuth into codex from a $20 OpenAI account for the last month. Works great. Anthropic as back up. Just used too many tokens. In another month or two, I’ll buy some mac with the new M5 and host a local model for the majority of tasks. Claude was a good bridge.

did you try hermes?

rip. i use ppq.ai as a fallback. guess it’ll be the primary for the immediate future

You got CloseClawed

If you have ollama running be sure to try Gemma 4 31B Just released by Google. Much better than expected and runs much lighter than expected. Running in full thinking mode at less than 20G usage on my mini. Performs high in tests. Tool calling, image reading, ocr, full thinking. Solid model for rep...

will nostr:nprofile1qyxhwumn8ghj7mn0wvhxcmmvqy3hwumn8ghj7mn0wd68yttjv4kxz7fwwpekvmm4dejxzarfdahzu6twvehsqg88sz3ry0pewggw2hd4yrhcukrpfxh3fjqw7k9d9e6af4eav4dzrg90f4qm survive this?? nostr:nprofile1qyt8wumn8ghj7etyv4hzumn0wd68ytnvv9hxgtcpzemhxue69uhks6tnwshxummnw3ezumrpdejz7qpq2rv5lskctqxxs2c8rf2zlzc7x...