anyone who milked google gemini free tier api just got slapped with new rate limiting for free tier. I had many accounts for individual web applications and boy did this just stop my work flow and some home assistant automations. Im sure my usage alone equates to nasa yearly budget, it was bound to happen.
I run my own local in Ollama on an Nvidia 4060. 8GB of memory but it could be better with more memory. Still low usage but it works.
You may want something bigger for a lot of usage…
Im sure my usage alone equates to nasa yearly budget, it was bound to happen.