You panicked about your token bill.
-
You panicked about your token bill. You grabbed Ollama. Now you're creating operational debt that's going to cost more than the tokens you saved.
Three real failure cases. One honest framework. Why rushing to self-hosted without discipline creates worse problems.
Read the full analysis: https://haunted.lighthouse.co.im/articles/the-token-escape-trap/?utm_source=mastodon
-
You panicked about your token bill. You grabbed Ollama. Now you're creating operational debt that's going to cost more than the tokens you saved.
Three real failure cases. One honest framework. Why rushing to self-hosted without discipline creates worse problems.
Read the full analysis: https://haunted.lighthouse.co.im/articles/the-token-escape-trap/?utm_source=mastodon
@alan A general question (I'm playing a little bit with local running AI, but keep tokens for "serious work") : how to evaluate a model ? How to evaluate regression ? How to get a "score" on the model performance ?
This includes the various benchmark I see sometimes, I have no idea how they are done (even in a high level view)
-
You panicked about your token bill. You grabbed Ollama. Now you're creating operational debt that's going to cost more than the tokens you saved.
Three real failure cases. One honest framework. Why rushing to self-hosted without discipline creates worse problems.
Read the full analysis: https://haunted.lighthouse.co.im/articles/the-token-escape-trap/?utm_source=mastodon
@alan You assume everyone who self-hosts is panicking and undisciplined, but many plan carefully to avoid exactly those failure cases.
-
R relay@relay.an.exchange shared this topic
