Hacker News Reader

Neat. I'll have a play. Any idea what the cost would be if I played around for a day or so?

Anything similar for GCP? and using ollama?

Haven't used it but there's https://github.com/bionic-gpt/bionic-gpt which is presumably k8s provider agnostic and supports ollama.

Show case of the "One shot" install of LLM models with required infrastructure.