vibecoding.tech_
vibecoding.tech_
🗺️AI mapGuide
🗺️AI mapGuide
NewsAI agent rankingComparisonsWhat vibecoding is
vibecoding.tech
© 2026
HomeMapGuideNews📬Newsletter

./open · api-costs

What the agents building this site cost

Agents have been building vibecoding.tech since July 1, 2026, and every message they send sits in the transcripts. The counter multiplies tokens by public prices and puts two totals side by side: what it would have cost via API and what was actually paid through subscriptions.

1 Jul 2026 - 11 Nov 2026 · 134 days · one counter from transcripts
97.2B
tokens burned
$74,219
would cost via API
$1,831
paid via subscriptions
×41
subscriptions cheaper than API
our price $19 per 1B tokens, via API $764
AgentModelSession
Claude Code$67.2K
Codex$7,028
Fable 5$29.5K
Opus 5.5$16.7K
Fable 5.1$11.5K
Opus 5$7,928
GPT-6 Astra$2,354
GPT-5.6 Sol$1,997
GPT-6.1 Sol$1,435
other models$2,773
Main session$55.6K
Subagents$18.6K
  • Fable 5$29,51839.8%
    18.3B tokens · $29,518 via API · $854 in subscriptions
  • Opus 5.5$16,74522.6%
    43.5B tokens · $16,745 via API · $251 in subscriptions
  • Fable 5.1$11,46715.5%
    13.1B tokens · $11,467 via API · $253 in subscriptions
  • Opus 5$7,92810.7%
    8.2B tokens · $7,928 via API · $151 in subscriptions
  • GPT-6 Astra$2,3543.2%
    1.5B tokens · $2,354 via API · $6 in subscriptions
  • GPT-5.6 Sol$1,9972.7%
    3.3B tokens · $1,997 via API · $130 in subscriptions
  • GPT-6.1 Sol$1,4351.9%
    6.1B tokens · $1,435 via API · $2 in subscriptions
  • Opus 4.8$1,3141.8%
    1.2B tokens · $1,314 via API · $28 in subscriptions
  • GPT-5.5$1,0391.4%
    1.3B tokens · $1,039 via API · $142 in subscriptions
  • and more 11
by daypeak $2,679
1 Jul 202611 Nov 2026
  • Every model message is counted once, even if a continued session copied the history into a new file.
  • Subagents are included, they have their own column in the stream.
  • Codex is counted by spend events, not by the file's total: a session fork carries the parent's total.
  • Prices from Anthropic's public price lists and the LiteLLM registry for OpenAI; where there is no price, an estimate based on GPT-5.6 Sol is used with a note; cache writes ×1.25 and ×2, reads at the model's cache price.
  • The subscription is spread across days and split between the day's models by their share of API dollars within their own vendor; the counter knows nothing about discounts or invoices and shows only this site.

Breakdown: why the subscription came out 35 times cheaper than API (snapshot Sep 27, 2026)The whole build live: machine, cost, selection