Local AI: Private Chats & Agentic Work Without Token Costs! #shorts

Transcript

1. 0:00That's one way local models are being used is just for local private chats for lack of a better term.
2. 0:08The other way that they're being used is once you start using these agentic harnesses like Chat GPT Codex or or Chat GPT work,
3. 0:19Claude code, um Gemini anti-gravity, Hermes, open claw, these things where you you're spinning up agents and you
4. 0:29you basically give it the equivalent of a goal and then it goes out and starts working on your behalf. The challenge with those platforms is
5. 0:39that they are token hogs. They just send massive amounts of context back and forth to open AI or or to Claude or to whoever
6. 0:50um without you really knowing what's being sent, but but they they're basically keeping these massive contexts
7. 0:56alive and doing these back and forth chats. So, what ends up happening is your monthly subscription is getting chewed up very very quickly.
8. 1:06So, if you want to do that kind of agentic work where the the things are working 24/7 and you don't want to blow through
9. 1:15your tokens, you know, your your 20 or $200 a month tokens, then using a local model will not burn any of their tokens.
10. 1:26>> Watch the full replay at community.thesalon.ai.