hailports articleshttps://www.hailports.com/articles/a base M4 mac mini with 24GB runs my local models and agents all day. real tok/s, real costs, every flag.i finally priced the work running on my minihttps://www.hailports.com/articles/priced-the-work-on-my-minihttps://www.hailports.com/articles/priced-the-work-on-my-mini2026-11-0316 days of local model traffic on a base M4 mac mini with 24GB, priced at API list rates: $165.61, or $310.52 per 30 days.41 tok/s from a 35B model on a base M4 mac mini with 24GBhttps://www.hailports.com/articles/41-tok-s-35b-model-base-m4-mac-minihttps://www.hailports.com/articles/41-tok-s-35b-model-base-m4-mac-mini2026-11-02the exact llama.cpp launch line, the benchmarks and the draft settings that took a base M4 24GB from 22.2 to a 41.3 tok/s peak.the draft looked fine until i checked the claimhttps://www.hailports.com/articles/the-draft-looked-fine-until-i-checked-the-claimhttps://www.hailports.com/articles/the-draft-looked-fine-until-i-checked-the-claim2026-11-02nine quality-gate rules for agent drafts at work: blocklists before judges, written rubrics, fail-closed scoring and holding claims with no source.how paid frontier LLM burn dropped 91% with a 24GB M4 mac minihttps://www.hailports.com/articles/cut-paid-llm-burn-94-percent-m4-mac-minihttps://www.hailports.com/articles/cut-paid-llm-burn-94-percent-m4-mac-mini2026-10-26moving routine agent work onto a $599 base M4 mac mini with 24GB took paid API burn from $42.50 a day to $3.90 a day.