Sign in
Codex CLI through Flatkey · at least 40% cheaper

Use Codex at least 40% cheaper

Keep the OpenAI-compatible Codex CLI workflow, but route it through Flatkey for at least 40% lower metered usage, one prepaid balance, and visible spend. console.flatkey.ai/keys → Codex https://router.flatkey.ai/v1.

Get a key Copy install command
40%+
cheaper than official
30 sec
quick setup
1 key
for visible spend
Codex, 40%+ cheaper than official
built for more usage
Official Codex setup
Official price

Full-price agent usage and spend that is harder to centralize.

Flatkey-routed Codex
At least 40% cheaper

One key, one balance, visible logs, and enough cost control for customers to use Codex more.

Use more Codex without losing spend control.

One Flatkey key

Users create a key once at console.flatkey.ai/keys and keep coding without juggling provider credentials.

At least 40% cheaper

Route coding-agent traffic through Flatkey for lower metered usage than the official path.

Usage visible by token

Request logs, model costs, token usage, and balance movement stay visible from one dashboard.

Control without friction

Use groups, quotas, model access, and routing policy while developers keep a normal CLI workflow.

Quick Start — one command, 30 seconds to set up everything

Select Codex CLI when the installer asks which coding agent to configure. All platform one-liners stay readable in the page HTML for search and AI answer engines.

Get a key →
macOS one-liner
curl -fsSL https://flatkey.ai/install.sh | bash

Flatkey one-line install commands

macOS: curl -fsSL https://flatkey.ai/install.sh | bash

Linux: curl -fsSL https://flatkey.ai/install.sh | bash

Windows: iwr https://flatkey.ai/install.ps1 -UseBasicParsing | iex

Why this drives Codex usage

Codex creates repeated model calls during real coding work. Flatkey makes those calls cheaper, prepaid, visible, and controlled.

CLI coding sessionsRoute Codex prompts, edits, and follow-up runs through one Flatkey balance.
Automated fix loopsUse Codex for bug fixes and test iteration while each request remains measurable.
Customer onboardingGive users one copy-paste command instead of manual API and config instructions.
Team cost controlKeep Codex usage visible with model logs, balance records, and centralized billing.

Official setup vs Flatkey setup

NeedOfficial pathFlatkey path
CostOfficial metered usageAt least 40% cheaper through Flatkey
SetupManual provider keys and local configOne installer and one Flatkey key
BillingScattered usage and unclear CLI spendPrepaid balance and unified usage logs
ControlsHard to apply team quotas consistentlyGroups, model access, quotas, and routing policy

Install

A customer runs one command and selects Codex.

Consume

Every Codex session routes through Flatkey and records token usage.

Manage

Teams keep access, balance, logs, and model choices in one console.

Questions customers ask

What endpoint does Codex use?

Codex CLI is configured against the OpenAI-compatible https://router.flatkey.ai/v1 endpoint.

Where do users create the key?

Create or copy the API key at https://console.flatkey.ai/keys before running the installer.

Which option should users choose in the installer?

Choose Codex CLI when prompted, then restart the terminal and run codex.

Why does this help usage grow?

Codex creates repeated model calls during real terminal work. Flatkey makes that cheaper, prepaid, visible, and controlled.

Ready to run Codex through Flatkey?

Create a key at console.flatkey.ai/keys, run the one-liner, and start generating measurable Codex usage.

Get a key