Do MyCodex API plans have fair-use limits?

Yes. Plans use large monthly allowances, fair-use rules, queueing, throttling, and add-on packs. This keeps heavy agent traffic from breaking capacity for everyone else.

What do I need to change in an OpenAI-compatible client?

Use the MyCodex API base URL, dashboard API key, and model name. Owner-side OpenAI, ChatGPT, Codex OAuth, prehook files, and private prompts stay server-side.

What happens when upstream capacity or a Codex Auth window is limited?

The response and usage ledger should expose a readable status such as usage-window limited, queued, throttled, or upstream error. Failed upstream capacity should not be charged as a successful call.

MyCodex API pricing

Free / Go / Plus / Pro / Business pricing

Name: MyCodex API
Brand: MyCodex API

MyCodex API: same-price anchors with about 2x visible usage

Subscription-first packaging with usage-ledger reconciliation underneath. Buyers see trial calls, monthly allowances, limit behavior, add-ons, and team seats; the system still accounts by request, tokens, model tier, capacity priority, and risk flags.

Get API Key View product logic

Codex subscription

Buy MyCodex API and agent coding usage through Go, Plus, Pro 5x, or Pro 20x plans.

API relay

Issue Dashboard API Keys from the same account for Cursor, Cline, Open WebUI, custom agents, and the OpenAI SDK.

Server-side secret boundary

Upstream tokens, local Codex config, prehook files, prompts, and private context stay server-side.

Plans and usage limits

Same-price official anchors with clearer endpoints, allowances, and limit explanations

The pricing page no longer makes legacy credits the primary purchase path. P0 emphasizes Free validation, Go/Plus individuals, Pro 5x/20x heavy agents, and Business team budgets.

Free

Connection validation

Free

10 successful calls or 1 trial day

Allowance: 10 successful calls or 1 trial day
RPM: Low RPM
Concurrency: Single validation lane
Overage: Upgrade to Go/Plus to continue

Validate MyCodex API, Codex workflows, and API setup.

Test the base_url, API key, model, and streaming.

Low concurrency; no carryover, transfer, or cash-out; upgrade prompt at limit.

Go

Light individual use

$8/month

Same price as ChatGPT Go, about 2x visible usage

Allowance: About 2x ChatGPT Go visible usage
RPM: Low RPM
Concurrency: Low individual concurrency
Overage: Temporary add-on at 25%-50% of monthly fee

For light MyCodex tasks and occasional agent usage.

Use with Cursor, Cline, Open WebUI, and small internal tools.

Single user; low concurrency; after fair-use, throttle, queue, or buy an add-on.

Plus

Recommended

Recommended individual plan

$20/month

Same price as ChatGPT Plus, API access + about 2x usage

Allowance: About 2x ChatGPT Plus visible usage
RPM: Standard RPM
Concurrency: Standard individual concurrency
Overage: Temporary add-on at 25%-50% of monthly fee

For daily MyCodex coding, automation, and agent workflows.

The same plan allowance powers Dashboard API Key calls.

Standard RPM; after fair-use, throttle/queue; add-ons available.

Pro 5x

Heavy coding agent use

$100/month

Same price as official Pro 5x, about 2x visible usage

Allowance: About 2x official Pro 5x visible usage
RPM: Higher RPM
Concurrency: Priority individual concurrency
Overage: Heavy add-on at 20%-40% of monthly fee

For longer MyCodex sessions and high-effort tasks.

Higher priority and RPM for heavy mixed agent/API usage.

Single user; fair-use or abnormal concurrency triggers throttling; bursts use add-ons.

Pro 20x

High-frequency agent / small production traffic

$200/month

Same price as official Pro 20x, highest individual capacity anchor

Allowance: About 2x official Pro 20x visible usage
RPM: High RPM
Concurrency: High-frequency agent concurrency pool
Overage: Heavy add-on at 20%-40% of monthly fee

For high-frequency agents, long tasks, and small production workflows.

Includes an agent dashboard, higher concurrency pools, and clearer usage reasons.

Still usable after fair-use with lower priority; heavy add-ons available.

Business

Teams and courses

$25/user/month monthly · $20/user/month annual

2-seat minimum, shared allowance pool and project keys

Allowance: About 2x Business visible usage per seat
RPM: Team pool RPM
Concurrency: Project-key isolated concurrency
Overage: Seat or shared pool add-ons

Team members use MyCodex and agent workflows by role.

Owners can create project keys, member budgets, and shared organization pools.

Add seats as needed; over organization allowance can auto-add or hard stop.

Add-ons and limit behavior

Go / Plus

Temporary add-on at 25%-50% of the current monthly fee

Avoid churn at the first overage while keeping the entry plan simple.

Pro 5x / Pro 20x

Heavy add-on at 20%-40% of the current monthly fee

Absorb agent spikes without pushing heavy users into low-tier capacity.

Business

Add seats or shared pool capacity separately

Teams buy management, reimbursement, isolation, and predictability.

Fair-use stated clearly

Every plan has fair-use, concurrency/RPM, queueing, and risk-control rules.

At the limit, show throttling, queueing, add-ons, or upgrade paths.

Heavy agents should not be allowed to exhaust low-tier capacity indefinitely.

Explainable billing

Reserve before a request and reconcile against real usage after completion.

Upstream errors, usage-window errors, and client disconnects must not be billed as successful calls.

Each charge should trace to request id, plan, model tier, and trace id.

Secret boundary

Customers receive only the dashboard API key, base_url, and model.

Owner-side OpenAI/Codex/OAuth tokens are never exposed to clients.

P0 uses metadata-only logging and does not store raw prompts/responses by default.

5-minute setup

Subscription and API usage share one account and allowance

Individuals can treat it as a Codex usage subscription. Developers and teams can connect the same allowance to an OpenAI-compatible API. Migration should only require changing base_url, api_key, and model; customers do not upload their own upstream key.

The free trial grants successful calls only, not a stockpiled wallet balance.

At the limit, show throttling, queueing, add-ons, or upgrade paths instead of silently dropping requests.

Metadata-only usage logs record request id, plan, usage, capacity state, and error code.

MyCodex API quickstart

Copy these MyCodex API values into Codex CLI-style tools, OpenAI-compatible SDKs, Cursor, Cline, or Open WebUI. The frontend endpoint validates your dashboard API key before forwarding to the private relay; the same account can also be used as a usage subscription.

base_url

https://codexcliapi.com/api/v1/codex

api_key

cps_live_your_dashboard_key

model

gpt-5.4-mini

Three-line curl

curl -N "https://codexcliapi.com/api/v1/codex/chat/completions" \
  -H "Authorization: Bearer cps_live_your_dashboard_key" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.4-mini","stream":true,"messages":[{"role":"user","content":"Say hello from the relay"}]}'

Cursor

Provider: OpenAI Compatible
Base URL: https://codexcliapi.com/api/v1/codex
API Key: cps_live_your_dashboard_key
Model: gpt-5.4-mini

Cline

API Provider: OpenAI Compatible
Base URL: https://codexcliapi.com/api/v1/codex
API Key: cps_live_your_dashboard_key
Model ID: gpt-5.4-mini
Streaming: enabled

Open WebUI

Connection type: OpenAI API compatible
API Base URL: https://codexcliapi.com/api/v1/codex
API Key: cps_live_your_dashboard_key
Model: gpt-5.4-mini

Use the dashboard API key issued here. Do not paste owner-side OpenAI Platform keys, ChatGPT/Codex OAuth tokens, private relay internal keys, local config, prehook files, or private prompts into client tools.

Codex API and Codex CLI access

Explore the Codex API keyword cluster: Codex CLI setup, Codex CLI API configuration, Codex API key access, pricing, fair-use limits, and OpenAI-compatible relay docs.

Codex API

Hub page for Codex API endpoint, pricing, limits, and setup.

Codex CLI

CLI-focused setup page for API key and base URL workflows.

Codex CLI API

Exact-match page for Codex CLI API key, URL, cost, and limits.

Codex API Key

High-intent page for API key, token, and access searches.

OpenAI Codex API

Comparison page for OpenAI Codex API compatible access intent.

Have feedback on our Codex API plans? Submit feedback or email us at [email protected]

FAQs

Codex API Frequently Asked Questions

Codex API, Codex CLI API, API key, pricing, limits, and OpenAI-compatible relay setup.

What is Codex API access on this site?

Codex API access means a buyer-facing OpenAI-compatible relay endpoint with a dashboard API key, base URL, model name, streaming chat completions, pricing tiers, and fair-use controls for Codex CLI-style agent workflows.

Can I use it with Codex CLI?

The setup is designed for tools that support custom OpenAI-compatible base URLs. Use the Codex API base URL, dashboard API key, and model name from the quickstart in Codex CLI-style tools, Cursor, Cline, Open WebUI, curl, or SDK clients.

How do I get a Codex API key?

Use the pricing and dashboard flow to get a customer-facing API key. That key belongs in client tools; private owner-side OpenAI, Codex, OAuth, prompt, prehook, and relay credentials remain server-side.

Is this the official OpenAI Codex API?

No. This site positions an OpenAI-compatible Codex API relay and storefront for developer workflows. It answers OpenAI Codex API search intent without claiming official OpenAI ownership.

Does Codex API access have limits?

Yes. Codex API access uses plan tiers, fair-use windows, concurrency, queueing, and capacity controls. Free access is for setup validation; heavy agent sessions should use paid capacity.

What is the Codex CLI API base URL?

The public relay path is shaped around /api/v1/codex. The exact full base URL depends on the production domain and is shown in the Codex relay quickstart.

Why is my Codex CLI API key not working?

Common causes include using the wrong key type, inactive billing, a revoked dashboard key, a wrong base URL, a mismatched model name, or a client that does not support custom OpenAI-compatible endpoints.

What keywords does the homepage target?

The homepage targets Codex API, Codex CLI, Codex CLI API, Codex API key, OpenAI Codex API, ChatGPT API key, OpenAI-compatible API, Codex API pricing, and Codex API limits.

Can I use the API with Cursor, Cline, or Open WebUI?

Yes, if the client supports OpenAI-compatible base URL configuration. The quickstart shows setup values for Cursor, Cline, Open WebUI, curl, and custom SDK clients.

How does pricing work for Codex API usage?

Pricing is presented as API-shaped access tiers for setup validation, daily individual use, heavy coding agents, high-frequency workflows, and team access, with fair-use and upgrade paths made explicit.

Need help with Codex API key setup, Codex CLI API configuration, pricing, or limits? Contact support.