UseRightAI
HomeModelsAsk AIComparePricingWhat's New
UseRightAICut through AI hype. Pick what works.

Independent AI model tracker. Live pricing, real benchmarks, zero vendor bias.

X (Twitter)LinkedInUpdatesContact

Compare

Opus 4.8 vs Opus 4.7Fable 5 vs Opus 4.8New AI Models 2026ChatGPT vs ClaudeGPT-4o vs Claude SonnetClaude vs GeminiDeepSeek vs ChatGPTMistral vs ClaudeGemini Flash vs GPT-4o MiniLlama vs ChatGPTAll comparisons →Build your own →

Best For

CodingWritingDevelopersProduct ManagersDesignersSalesBest Cheap AIBest Free AI

Pricing & Data

API Token PricingCost per TaskPrice HistoryBenchmark ScoresPrivacy & SafetySubscription PlansPlan Usage LimitsCost CalculatorWhich AI is Cheapest?Cheapest AI APIs

Company

About UseRightAIContactWhat ChangedAll ModelsGuidesEditorial PolicyDisclosuresPrivacy PolicyTerms of Service

© 2026 UseRightAI. Independent · Free forever · Not affiliated with any AI provider.

Affiliate links are clearly labeled. See disclosures.

HomeAI Coding AgentsOpenAI Codex

Coding Agent Guide

OpenAI Codex: Pricing, Limits and Plans (October 2026)

OpenAITerminal · IDE · Desktop app · Cloud agentVerified October 13, 2026

Quick answer

Codex comes with every ChatGPT plan. On Plus ($20/month) OpenAI estimates 15–160 local messages per 5-hour window on GPT-6.1 Sol and 5–45 on GPT-6 Astra, with weekly limits on top. The Pro tiers ($100, $200 or $500 a month) have no 5-hour limit, and the $500 tier adds Ultrafast mode. Past your limit, Plus and Pro users can buy credits.

Models: OpenAI models only — GPT-6.1 Sol and GPT-6 Luna on Plus, plus GPT-6 Astra; GPT-5.5 leaves Codex on October 14, 2026.

Plans and what they include

Official rows come from OpenAI's own pricing page; reported rows from press or trackers we could not confirm there.

PlanPriceWhat you get
Freeofficial$0GPT-6 Luna at standard speed in the desktop app, subject to rollout. No published limits.GPT-6 Luna at standard speed in the desktop app, subject to rollout. No published limits.
Goofficial$8/moSame GPT-6 Luna access as Free. No published limits.Same GPT-6 Luna access as Free. No published limits.
Plusofficial$20/moGPT-6.1 Sol and GPT-6 Luna. Per 5 hours: GPT-6 Astra 5–45, GPT-6.1 Sol 15–160, GPT-6 Sol 15–150, GPT-6 Luna 350–3,000 local messages; weekly limits may apply.GPT-6.1 Sol and GPT-6 Luna. Per 5 hours: GPT-6 Astra 5–45, GPT-6.1 Sol 15–160, GPT-6 Sol 15–150, GPT-6 Luna 350–3,000 local messages; weekly limits may apply.
Proofficial$100, $200 or $500/moNo five-hour limit; higher tiers get more included usage. Ultrafast mode on the $500 tier.No five-hour limit; higher tiers get more included usage. Ultrafast mode on the $500 tier.
Businessofficial$25/user/mo ($20 billed annually)2+ users. Roughly Plus-level limits per seat; workspace credits for extra usage.2+ users. Roughly Plus-level limits per seat; workspace credits for extra usage.
Enterprise / EduofficialContact salesFlexible-pricing workspaces have no fixed rate limits.Flexible-pricing workspaces have no fixed rate limits.
API keyofficialPay per tokenGPT-6.1 Sol $2/$10, GPT-6 Luna $0.10/$0.50, GPT-6 Astra $10/$50 per million tokens.GPT-6.1 Sol $2/$10, GPT-6 Luna $0.10/$0.50, GPT-6 Astra $10/$50 per million tokens.

How billing and limits work

  • Plus and Standard Business are metered in 5-hour windows; OpenAI publishes estimated message ranges per model rather than fixed counts.
  • Local messages and cloud chats share the same allowance.
  • Fast mode uses included usage 2.5x faster and Ultrafast 8x faster on GPT-6 Astra and GPT-6.1 Sol.
  • Plus and Pro users who hit the limit can buy credits instead of upgrading. OpenAI prices credits per million tokens by model (GPT-6.1 Sol: 50 input / 250 output credits) but does not publish a dollar price per credit on the Codex page.
  • Pro plans have no 5-hour limit at all; weekly limits may still apply.

Best for

  • Anyone already paying for ChatGPT Plus — Codex is included.
  • Teams on OpenAI's stack, where GPT-6.1 Sol gives near-Astra coding at Sol prices.
  • Very heavy users who want no 5-hour cap, on a Pro tier.

Watch out for

  • OpenAI models only.
  • The published message estimates are wide ranges — real usage depends heavily on task size and model.
  • Pro pricing now spans three tiers ($100/$200/$500); check which one includes the speed mode you need.

What changed, and when

October 14, 2026 official

GPT-5.5 retires from ChatGPT, ChatGPT Work and Codex on all plans (the API is not affected).

September 29, 2026 official

GPT-6.1 Sol launches at DevDay for paid plans through ChatGPT Work and Codex.

September 22, 2026 official

GPT-6 Sol and GPT-6 Luna launch.

September 3, 2026 official

GPT-6 Astra launches and becomes available in Codex.

FAQ

Is OpenAI Codex free?

Codex is included in every ChatGPT plan, including Free and Go, which get GPT-6 Luna in the desktop app as it rolls out. Serious use starts with Plus at $20/month, which adds GPT-6.1 Sol.

How many Codex messages do I get on Plus?

OpenAI estimates 15–160 local messages per 5-hour window on GPT-6.1 Sol, 15–150 on GPT-6 Sol, 5–45 on GPT-6 Astra and 350–3,000 on GPT-6 Luna, with weekly limits on top. The range is wide because bigger tasks use more.

What do the Codex Pro tiers cost?

Pro starts at $100/month, with tiers at $100, $200 and $500. All have no 5-hour limit; the $500 tier adds Ultrafast mode.

Can I use Codex with an API key?

Yes. Usage then bills per token at API rates — GPT-6.1 Sol is $2 per million input tokens and $10 per million output.

Sources

  • OpenAI — Codex / ChatGPT Work pricing and usage limits

Coding-agent pricing changed for almost every vendor in 2026, often without notice. Everything above was checked against the linked sources on October 13, 2026; if you spot something outdated, email hello@userightai.com.

Compare with

All coding agents side by side →Chatgpt Plus LimitsGpt 6 1 SolGpt 6 AstraClaude Opus 5 5 Vs Gpt 6 1 Sol

Newsletter

Get an email when coding-agent pricing changes

Vendors change plans and limits quietly. We re-verify and log every change, dated.

No spam. Useful updates only. Affiliate disclosures always clearly labeled.