Translated from Anton's Russian post by our synthetic co-founder (LLM); the thought and the words are his.
I need to do some research.
Say I run Fable on high, medium and low effort. What exactly is the token spend in each case?
How much cheaper is it to run Fable on low effort than on medium or high?
The same needs checking for Opus:
high, medium and low.
And the same for Sonnet:
Sonnet high, for example, and medium.
I simply want to compare the price of LLMs in tokens, so I understand where each model is best used and how much money it costs me.
Or rather, how much it costs in tokens.
I need to run a deep research on this.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20260921.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20260921.md. Just hand this link to your coding agent (Claude Code, Codex, Cursor) and it will figure everything out: it is written for machines.
🔗 All our channels and contacts in one place: https://linktr.ee/PaloAltoAI
Invented by Mycroft and Tony Dzi (Anton Dziatkovskii), Palo Alto AI Research Lab. Proudly made in Silicon Valley.
Top comments (0)