SERIES Understanding and Managing the AI Agent Footprint: A How-To Series
Understanding and Managing the AI Agent Footprint: A How-To Series

What is the Understanding and Managing the AI Agent Footprint Series?

AI agents are now integrated directly into development tools, financial software, and other sensitive workflows. But there is a gap between what agents are capable of and what users know about what they actually do on a device. This series provides practical guidance on how to understand, monitor, and manage the footprint agents leave on your system, so you can work with them with greater accountability and confidence.

This section focuses on understanding why token costs are higher than expected and how to reduce unnecessary spending and includes:

How Much Does Codex Cost? (June 2026)

OpenAI's Codex is an AI coding agent integrated into ChatGPT that runs tasks in a cloud sandbox — reading repositories, writing code, running tests, and producing pull requests autonomously. Pricing changed in April 2026 when OpenAI switched from per-message billing to token-based billing.

Quick Answer: Codex is included in ChatGPT Plus ($20/month), Pro ($100 or $200/month), Business (~$30/user/month), and Enterprise plans. There is no separate Codex subscription. API access bills per token — GPT-5.1 Codex mini is the most cost-efficient option for routine tasks, GPT-5.4-Mini handles mid-complexity work, and GPT-5.5 is the premium tier for complex multi-step engineering. Average developer spend runs $100–$200 per developer per month.

What is included in each ChatGPT plan?

Codex agent access comes with existing ChatGPT subscriptions. You do not need a separate product or credit purchase to use it.

Plan Monthly Cost Codex Access
Free $0 Limited Codex Mini access with daily cap
Plus $20 Usage-limited; purchase additional credits if exhausted
Pro $100 5x usage vs. Plus; all Pro features
Pro $200 20x usage vs. Plus; highest subscription limits
Business ~$30/user Org-wide access, admin controls, data not used for training
Enterprise Custom No fixed rate limits; scales with purchased credits

Starting April 2, 2026, Codex usage bills based on API token consumption rather than message counts. Additional credits can be purchased if you exhaust plan limits.

What are the Codex API token rates?

API access bills per million tokens with separate input and output pricing:

ModelInput per 1M tokensOutput per 1M tokens
GPT-5.1-Codex-Mini$0.25$2.00
GPT-5.3-Codex$1.75$14.00
GPT-5.4 Mini$0.75$4.50
GPT-5.4$2.50$15.00
GPT-5.5$5.00$30.00

Pricing via OpenRouter. Rates update automatically.

For most coding tasks, GPT-5.4-Mini offers the best cost-to-performance ratio. GPT-5.3-Codex and GPT-5.4 handle complex multi-file reasoning and agentic loops, while GPT-5.5 is reserved for the highest-stakes tasks where maximum capability matters.

Cached input tokens — context that repeats across turns in a session — cost approximately 10% of the standard input rate. In agentic coding sessions where repository context is carried across many turns, caching is the most impactful cost optimization available.

How much does Codex cost in practice?

Published data from enterprise deployments shows average spend of $100–$200 per developer per month, with significant variance based on workflow intensity.

In ChatGPT auth mode (the default), Codex usage draws from your plan's included limits at no extra charge. If you switch to API key mode, you are billed at standard token rates.

The Codex Pro 2x promotional pricing ended May 31, 2026. Users on the Pro plan at $200/month saw effective capacity halve from June 1 at the same price.

What are common mistakes to avoid?

  • Using GPT-5.4 or GPT-5.5 via the API for tasks that GPT-5.1 Codex mini or GPT-5.4-Mini handles adequately
  • Not enabling prompt caching for sessions where repository context repeats across many turns
  • Switching to API key mode without tracking per-session token costs
  • Running Codex in a loop without iteration limits, which can exhaust plan credits rapidly
  • Not purchasing additional credits before hitting hard limits in production workflows

Find Out Where Your Token Budget Is Actually Going

Most teams track how many tokens their agents use. Few know whether those tokens produced useful work. AgentGuard360 Cost Intelligence runs as a background service — no SDK, no instrumentation required — and generates an efficiency grade (A–F) calibrated against peers running the same agent type. The report breaks waste down by driver: prompt overhead, retry loops, and model selection. Each line shows the token cost of the inefficiency and the estimated 7-day savings if fixed. It also surfaces cheaper model alternatives for tasks where you are overpaying on capability you do not need.

Coming Soon

Frequently Asked Questions

What is included in each ChatGPT plan?

Codex agent access comes with existing ChatGPT subscriptions. You do not need a separate product or credit purchase to use it.

Plan Monthly Cost Codex Access
Free $0 Limited Codex Mini access with daily cap
Plus $20 Usage-limited; purchase additional credits if exhausted
Pro $100 5x usage vs. Plus; all Pro features
Pro $200 20x usage vs. Plus; highest subscription limits
Business ~$30/user Org-wide access, admin controls, data not used for training
Enterprise Custom No fixed rate limits; scales with purchased credits

Starting April 2, 2026, Codex usage bills based on API token consumption rather than message counts. Additional credits can be purchased if you exhaust plan limits.

What are the Codex API token rates?

API access bills per million tokens with separate input and output pricing:

For most coding tasks, GPT-5.4-Mini offers the best cost-to-performance ratio. GPT-5.3-Codex and GPT-5.4 handle complex multi-file reasoning and agentic loops, while GPT-5.5 is reserved for the highest-stakes tasks where maximum capability matters.

Cached input tokens — context that repeats across turns in a session — cost approximately 10% of the standard input rate. In agentic coding sessions where repository context is carried across many turns, caching is the most impactful cost optimization available.

How much does Codex cost in practice?

Published data from enterprise deployments shows average spend of $100–$200 per developer per month, with significant variance based on workflow intensity.

In ChatGPT auth mode (the default), Codex usage draws from your plan's included limits at no extra charge. If you switch to API key mode, you are billed at standard token rates.

The Codex Pro 2x promotional pricing ended May 31, 2026. Users on the Pro plan at $200/month saw effective capacity halve from June 1 at the same price.

What are common mistakes to avoid?
  • Using GPT-5.4 or GPT-5.5 via the API for tasks that GPT-5.1 Codex mini or GPT-5.4-Mini handles adequately
  • Not enabling prompt caching for sessions where repository context repeats across many turns
  • Switching to API key mode without tracking per-session token costs
  • Running Codex in a loop without iteration limits, which can exhaust plan credits rapidly
  • Not purchasing additional credits before hitting hard limits in production workflows