Codex BetterToken setup guide

Connecting Codex to BetterToken requires a custom model provider and API Key. The base_url must be https://www.bettertoken.ai/v1, and the protocol is OpenAI Responses.

Prerequisites

Node.js installed
BetterToken API Key (register here (https://www.bettertoken.ai/register))

Clear OpenAI-related environment variables before setup so they do not override the settings in config.toml or auth.json.Clear these variables:

OPENAI_API_KEY
OPENAI_BASE_URL

Run on macOS / Linux:

unset OPENAI_API_KEY
unset OPENAI_BASE_URL

Run in Windows PowerShell:

Remove-Item Env:OPENAI_API_KEY -ErrorAction SilentlyContinue
Remove-Item Env:OPENAI_BASE_URL -ErrorAction SilentlyContinue
[Environment]::SetEnvironmentVariable("OPENAI_API_KEY", $null, "User")
[Environment]::SetEnvironmentVariable("OPENAI_BASE_URL", $null, "User")

Install Codex

npm install -g @openai/codex

Setup

Automatic setup
Configure via CC Switch
Manual setup

You can run the BetterToken automatic setup script directly. It only writes the BetterToken provider configuration and does not replace the installation prerequisites above. The script requires Node.js. If you do not pass an API Key or model explicitly, the script prompts for them when the tool supports interactive input.Follow the step for your operating system. You only need one of them.

If you use macOS / Linux

curl -fsSL https://www.bettertoken.ai/install-codex-provider.sh | bash

If you use Windows

iwr https://www.bettertoken.ai/install-codex-provider.ps1 -OutFile "$env:TEMP\install-codex-provider.ps1"; powershell -ExecutionPolicy Bypass -File "$env:TEMP\install-codex-provider.ps1"

Copy a model ID from the model plaza

Open the BetterToken model plaza (https://www.bettertoken.ai/pricing) and copy a currently available model ID from the Codex group. Replace every YOUR_CODEX_MODEL_ID below with that value.

Confirm the config file locations

Manual setup requires both the Codex config.toml and auth.json files.

System	`config.toml`	`auth.json`
macOS / Linux	`~/.codex/config.toml`	`~/.codex/auth.json`
Windows	`C:\Users\YOUR_USER\.codex\config.toml`	`C:\Users\YOUR_USER\.codex\auth.json`

Edit the Codex config file

Open config.toml. Create it if it does not exist.

This config only applies when you manually edit ~/.codex/config.toml. If you configure Codex through CC Switch, follow the CC Switch guide exactly. Do not copy this block to replace configuration managed by CC Switch.

Add the following config:

model_provider = "bettertoken"
model = "YOUR_CODEX_MODEL_ID"
model_reasoning_effort = "high"
disable_response_storage = true

[model_providers.bettertoken]
name = "BetterToken"
base_url = "https://www.bettertoken.ai/v1"
wire_api = "responses"
requires_openai_auth = true
model_context_window = 1000000
model_auto_compact_token_limit = 900000

Write the Codex auth file

Open auth.json. Create it if it does not exist.Write the following content, replacing YOUR_API_KEY with your BetterToken API Key:

{
  "OPENAI_API_KEY": "YOUR_API_KEY"
}

Restart Codex in the terminal

After changing config.toml or auth.json, stop the current Codex process and start it again in a new terminal session.

Start Codex

codex

Config parameter reference

Parameter	Description
`model_provider`	The default model provider ID. It must match `bettertoken` in `[model_providers.bettertoken]`
`model`	The default model ID Codex uses after startup. Use a model copied from the Codex group in the model plaza
`model_reasoning_effort`	Model reasoning effort. Set it to `low`, `medium`, or `high` as needed
`disable_response_storage`	Disables response storage. Keep this set to `true`
`name`	Provider display name. You can keep it as `BetterToken`
`base_url`	BetterToken Codex / OpenAI-compatible protocol endpoint. It must include `/v1`
`wire_api`	The API protocol used by Codex. Keep this set to `responses`
`requires_openai_auth`	Tells Codex to read the API Key from its OpenAI auth file. Keep this set to `true`

1M context config

If you use a long-context model from the Codex group, keep these two settings:

model_context_window = 1000000
model_auto_compact_token_limit = 900000

Parameter	Purpose
`model_context_window`	Sets the available context window to `1,000,000` tokens
`model_auto_compact_token_limit`	Triggers compaction before the context reaches the hard limit, reducing the chance that a conversation runs out of context

If you write the config through CC Switch, prefer the config generated by CC Switch. Only edit these values yourself for manual setup or troubleshooting.

Verify the setup

After startup, send a test message. A normal reply means the connection is working. If something fails, check:

base_url is https://www.bettertoken.ai/v1
wire_api is still responses
model is a model ID copied from the Codex group
OPENAI_API_KEY in ~/.codex/auth.json is your BetterToken API Key
No old OPENAI_API_KEY or OPENAI_BASE_URL variables are interfering

​Prerequisites

​Install Codex

​Setup

​Config parameter reference

​1M context config

​Verify the setup

Prerequisites

Install Codex

Setup

Config parameter reference

1M context config

Verify the setup