OpenAI Codex Preisgestaltung: Was die CLI wirklich kostet 2026
OpenAI Codex CLI is a free, open-source tool — but it sends every prompt to the OpenAI API, which charges per token. One image compression task burned 7,675 tokens: between 3 and 11 cents. The default model, codex-mini-latest, costs $1.50 per million input tokens and $6.00 per million output tokens. Pointing Codex at RunAPI instead cuts those rates by 50%, with no other changes to your workflow.
Wie viel kostet OpenAI Codex?
The Codex CLI itself is free to download and run. The cost comes from API calls: every prompt, file read, and code generation goes to the OpenAI API. The default model is codex-mini-latest, priced at $1.50/1M input tokens and $6.00/1M output tokens through OpenAI directly. A moderate coding session — roughly 8M input and 2M output tokens per month — runs about $24 with OpenAI. The same usage through RunAPI costs $12. RunAPI is OpenAI-compatible, so you change two environment variables and nothing else. Tip: run /compact during long sessions to free up context and stop burning tokens on conversation history.
Die CLI selbst hat keine Abonnementgebühr. Sie zahlen nur für die während jeder Coding-Sitzung verbrauchten API-Tokens.
codex-mini-latest Eingabe kostet $1,50/1 Mio. Tokens, Ausgabe kostet $6,00/1 Mio. Tokens über die OpenAI-API.
RunAPI liefert dasselbe Modell zu 50 % Rabatt. Setzen Sie zwei Umgebungsvariablen und Codex routet automatisch über RunAPI.
Zahlen Sie nur für das, was Sie verwenden. Keine monatliche Sitzgebühr. Kein Mindestbetrag. Die Abrechnung stoppt, sobald Sie aufhören zu coden.
Was wird als Token gezählt?
Every message Codex sends to the API includes your prompt, the full context of files it has read, any tool results, and conversation history. That combined input is tokenized (roughly 750 words per 1,000 tokens). The model's response — edited code, explanations, next steps — counts as output tokens. Both directions are billed separately. Codex caches nothing between sessions, so a fresh session re-reads the same files and bills that context again. One real example from a developer testing image compression — a single 5 MB file task burned 7,675 tokens, or roughly 3 to 11 cents. Not very cheap for one file, but most coding sessions stay well under a dollar each.
codex-mini-latest Preisgestaltung: OpenAI vs. RunAPI
Preise in USD pro Million Tokens. RunAPI-Tarife sind 50 % der OpenAI-Listenpreise.
Was kostet ein typischer Monat?
Basierend auf codex-mini-latest zu $1,50/$6,00 (OpenAI) und $0,75/$3,00 (RunAPI) pro 1 Mio. Tokens.
So verwenden Sie Codex CLI mit RunAPI
Codex CLI installieren
RunAPI-Konto erstellen
Umgebungsvariablen setzen
Codex ausführen
Das Modell überprüfen
Häufige Fragen zur Codex-Preisgestaltung
Ist OpenAI Codex CLI kostenlos nutzbar?
Die CLI selbst kann kostenlos heruntergeladen und installiert werden. Es gibt keine Abonnementgebühr. Jede Eingabeaufforderung, die Codex sendet, erreicht jedoch die OpenAI-API (oder RunAPI), die pro Token berechnet. Das Tool ist kostenlos; die API-Aufrufe sind es nicht. Die meisten Entwickler zahlen je nach Intensität der Nutzung zwischen $3 und $30 pro Monat.
"Quota exceeded" — I'm already on a paid plan. Why?
Codex CLI works two ways: through a ChatGPT subscription (Plus at $20/month or Pro at $200/month) or through your own OpenAI API key with pay-per-token billing. The subscription path gives you a monthly quota — once exhausted, you see "quota exceeded" errors even on a paid plan. The API key path means you pay only for what you use, with no monthly floor. Heavy users often find the $200 Pro subscription impossible to exhaust; lighter users get better value from the pay-per-token API key path.
Which model effort level should I use to keep costs down?
Codex CLI includes three reasoning effort levels selectable with the /model command: medium, high, and extra high. Medium is the cheapest and recommended for daily use — it saves a lot of tokens and handles most tasks well. High and extra-high effort produce better results on complex tasks but burn more tokens quickly. ChatGPT Plus ($20) users should stick to medium; Pro ($200) subscribers can use high and extra-high without worrying about burning through quota.
How does RunAPI work with Codex CLI?
RunAPI is an OpenAI-compatible proxy. Codex reads the OPENAI_BASE_URL and OPENAI_API_KEY environment variables to find its API endpoint. Setting these to RunAPI values routes all Codex traffic through RunAPI instead of OpenAI. The request and response format is identical — Codex behaves the same, but you pay 50% less per token.
Do I need an OpenAI account to use Codex with RunAPI?
No. When you route Codex through RunAPI, your requests go to RunAPI's infrastructure, not OpenAI directly. You create a RunAPI account, generate an API key there, and set that key in your environment. No OpenAI account, no OpenAI API key, and no OpenAI billing are required.
Which models can I use with Codex CLI through RunAPI?
RunAPI supports codex-mini-latest as well as other LLMs including GPT-4o, Claude Sonnet, Gemini Pro, and more. You switch models with the --model flag or /model command in Codex. All RunAPI models are priced at 50% off the respective provider's list price.
Does redirecting Codex to RunAPI change how it behaves?
No. The only difference is which server processes the API request. Codex's file reading, code execution sandbox, approval modes, and MCP tool integrations all continue to work as normal. The request and response format RunAPI uses is identical to OpenAI's API, so Codex cannot tell the difference.
How do I stop burning so many tokens in a long Codex session?
Use the /compact command during long sessions — it summarizes the conversation to free up context and cuts the tokens you're burning on history. Also run /status to see your current token usage. For day-to-day work, stick to medium reasoning effort. For one-off tasks, use non-interactive mode (codex -q "your task") instead of an open session to avoid accumulating context.
Mehr entdecken
Vollständiges OpenAI-Modell-Lineup – GPT-5.4, o3 und mehr zu 50% Rabatt.
Kostenvergleich: Claude Code mit API-Abrechnung vs. Max-Abo.
Funktions- und Kostenvergleich der beiden führenden KI-Coding-Tools.
Jedes Modell mit einem einzigen Befehl im Terminal ausführen.
Setzen Sie zwei Umgebungsvariablen und jede Codex-Sitzung läuft über RunAPI zu 50 % der OpenAI-Tarife. Kein Abonnement, kein Mindestbetrag — Sie zahlen nur für verwendete Tokens.