Integration tutorial

Connect Codex

Point the built-in OpenAI provider at DiscountedTokens using the Responses API.

Buy or create a key

Use guest checkout or create an account key.

Set the base URL

https://discountedtokens.com/v1

Send a small test

Verify the route before moving production traffic.

~/.codex/config.toml

model = "gpt-5.6-sol" openai_base_url = "https://discountedtokens.com/v1"

Run and verify

export OPENAI_API_KEY="YOUR_DISCOUNTEDTOKENS_KEY" codex

Troubleshooting

Use the user-level config file. Codex ignores provider and base-URL settings in project-local config.
Open /status and confirm the configured model and provider URL.
Basic Responses text routing is certified here. Tool calls, background agents, and streaming must be tested for your Codex version before production use.
If authentication fails, confirm OPENAI_API_KEY is visible in the shell that launches Codex.

Configuration source: official documentation. DiscountedTokens is an independent gateway and is not endorsed by the client vendor.

Get an API key → All tutorials

Successful API requests are billed at the displayed token rates, metered to a fraction of a cent. Charges accumulate and are deducted from your prepaid balance in whole US cents, so a small request is not rounded up to a minimum charge.