The AI coding cost & usage library
Straight answers on what your AI coding agents cost and how close you are to your limits — model pricing, calculators, rate-limit explainers, and honest comparisons. Claude Code, Codex, Grok, and Muse Spark are tracked today; 59 more providers are on the roadmap.
Pricing accuracy
Grok 4.6 stays in the total
Grok Build records Grok 4.6 under its own internal model ID. JP maps that usage to xAI's canonical Grok 4.6 rates — $2 input, $0.50 cached input, and $6 output per million tokens — so the session is counted instead of omitted as unpriced.
Value benchmark
Best bang for the buck
Ranked from JP the Cat's bundled model pricing table using a standard coding-agent session — 500K input, 100K output, and 2000K cache-read tokens — then divided by qualitative effort (high / medium / low). Higher effort at lower session cost wins. Codex, Gemini, Grok, and Muse Spark figures are notional list price, not subscription spend.
Best medium effort
Composer 2.5
$0.60 / session · notional
Best low effort
Gemini 2.5 Flash-Lite
$0.11 / session · notional
| # | Model | Effort | Session cost | Basis |
|---|---|---|---|---|
| 1 | GPT-5.6 LunaOpenAI (GPT-5.x / Codex) | High | $0.26 | Notional |
| 2 | Gemini 2.5 Flash-LiteGoogle Gemini | Low | $0.11 | Notional |
| 3 | Composer 2.5xAI (Grok) | Medium | $0.60 | Notional |
| 4 | GPT-5.1-Codex-MiniOpenAI (GPT-5.x / Codex) | Low | $0.37 | Notional |
| 5 | Grok 4.3xAI (Grok) | High | $1.27 | Notional |
| 6 | Muse Spark 1.1Meta (Muse Spark) | High | $1.35 | Notional |
| 7 | Gemini 2.5 FlashGoogle Gemini | Low | $0.46 | Notional |
| 8 | Grok Build 0.1xAI (Grok) | Medium | $1.10 | Notional |
See blended $/MTok or tokens-to-dollars to model your own mix.