Codex

The Codex CLI and the ChatGPT desktop app both read ~/.codex/config.toml, so connecting one connects the other. Close open Codex sessions and quit the ChatGPT desktop app before connecting or running off. In a terminal, Nexus CLI waits for them to close; in a script, it exits with an error. --force writes anyway, which isn’t recommended. Generally available OpenAI Responses Nexus Auto Native web search

Quick start

Install and sign in

curl -fsSL https://nexus.fireworks.ai/install.sh | bash

Requires Node.js 18 or later, git, and bash. On Windows, run it from Git Bash; PowerShell breaks the script’s line endings.

nexus login

Connect Codex

nexus codex
nexus codex status

Restart and pick a model

Start a new Codex session. Codex opens on Nexus Auto (auto). Run nexus codex again at any time. It keeps the model you’re on.

/model lists Codex’s OpenAI models first, then an Auto row for Nexus Auto, then the other Fireworks models. Set effort in that picker.

Use Nexus Auto

Codex keeps its own sign-in. Pick ChatGPT, a device code, or an API key when it asks, or run codex login. Your Fireworks key pays for Fireworks models. Your Codex sign-in pays for OpenAI models; see Who pays for OpenAI model turns for the details.

Nexus Auto (auto) is supported on the Responses API, and Codex defaults to it.
On Codex, auto picks between GPT Sol and Fireworks open models for each turn. See how auto picks its closed model.
With only a Fireworks key, auto uses GPT Sol if you have an OpenAI Provider Key, otherwise Claude Opus if you have an Anthropic Provider Key, otherwise open models.
Codex doesn’t send an Anthropic key, so Anthropic model turns need an Anthropic Provider Key.
With manual setup, Codex can send x-anthropic-api-key and x-openai-api-key headers.

Who pays for OpenAI model turns

With Nexus CLI, your Codex sign-in decides who pays for OpenAI model turns:

Codex sign-in	Who pays for OpenAI model turns
ChatGPT	Your ChatGPT plan. Older models, such as `gpt-5.6-terra`, use an OpenAI Provider Key instead
OpenAI API key	That OpenAI key
Fireworks key only	An OpenAI Provider Key. Without one, an OpenAI model you pick fails with `no usable openai credential`

For OpenAI model turns, Nexus Auto uses an x-openai-api-key header first, then your Codex sign-in, then an OpenAI Provider Key.
Your ChatGPT plan doesn’t cover gpt-5.5, gpt-5.6-terra, or gpt-5.6-luna. Those turns need an OpenAI Provider Key.
Nexus CLI doesn’t hand a saved OpenAI key to Codex. To pay with it, sign Codex in with it: nexus key export --openai | codex login --with-api-key.
No browser on the machine? codex login --device-auth signs in with ChatGPT through a one-time code.

MCP and tools

Your servers stay in [mcp_servers] in ~/.codex/config.toml. MCP and app connector tools load through Codex’s tool search on every model.

# ~/.codex/config.toml, next to the settings Nexus CLI adds
[mcp_servers.docs]
command = "npx"
args = ["-y", "your-mcp-server"]

Native web search is supported while connected through Nexus CLI. Fireworks runs the search server-side and returns the results, with no MCP server or search service to set up.

Troubleshooting

Nothing changed after --model? Start a new or resumed session. Codex loads a new model only then. Upgrading from an earlier version? See Upgrading from earlier versions. When you resume an old session, select the provider explicitly:

codex resume <id> -c model_provider="fireworks-ai"

MiniMax models don’t work on Codex. Codex may insert assistant messages between tool_calls and tool_results, which MiniMax templates reject. Nexus CLI leaves MiniMax out of the Codex model list and rejects --model minimax-.... Use a Chat Completions harness such as OpenCode for MiniMax.

Manual setup (advanced)

Add a Fireworks provider

~/.codex/config.toml

model_provider = "fireworks-ai"
model = "glm-latest"

[model_providers.fireworks-ai]
name = "Fireworks"
base_url = "https://api.fireworks.ai/inference/v1"
wire_api = "responses"
env_key = "FIREWORKS_API_KEY"

Run Codex

codex

Codex answers from glm-latest.

Without a model catalog, Codex warns that it has no metadata for Fireworks models and uses fallback limits. The catalog also turns on MCP tool search and runs auto-review on your chosen model. Nexus CLI writes that catalog for you. To pay for OpenAI model turns with your Codex sign-in, as Nexus CLI does, replace the env_key line with these two:

~/.codex/config.toml

requires_openai_auth = true
env_http_headers = { "x-fireworks-api-key" = "FIREWORKS_API_KEY" }

Codex then sends its own sign-in, and reads your Fireworks key from the environment. See what changes without the Nexus CLI.

What the Nexus CLI writes

~/.codex

config.toml

fireworks-model-catalog.json

See what Nexus CLI writes

~/.codex/config.toml

model_provider = "fireworks-ai"
model_catalog_json = "~/.codex/fireworks-model-catalog.json"
model = "auto"

[model_providers.fireworks-ai]
name = "Fireworks"
base_url = "https://api.fireworks.ai/inference/v1"
wire_api = "responses"
requires_openai_auth = true
http_headers = { "X-Title" = "Codex", "HTTP-Referer" = "nexus/v<version>", "x-fireworks-api-key" = "fw_..." }

requires_openai_auth = true makes Codex send its own sign-in, and the Fireworks key travels in x-fireworks-api-key. Nexus CLI writes the same provider table for every model and sends no Anthropic or OpenAI key.Nexus CLI keeps your other TOML settings, including [mcp_servers], saves the file with mode 0600, and writes a model catalog so Codex knows each model’s context window and capabilities. The catalog also turns on Codex’s MCP tool search for every model, so MCP and app connector tools load on demand, and makes Codex auto-review run on the model you picked.

Get started

Nexus CLI

Model Orchestration

Harness Support

Nexus settings

More

Quick start

Use Nexus Auto

Who pays for OpenAI model turns

MCP and tools

Troubleshooting

Manual setup (advanced)

What the Nexus CLI writes

​Quick start

​Use Nexus Auto

​Who pays for OpenAI model turns

​MCP and tools

​Troubleshooting

​Manual setup (advanced)

​What the Nexus CLI writes

Quick start

Use Nexus Auto

Who pays for OpenAI model turns

MCP and tools

Troubleshooting

Manual setup (advanced)

What the Nexus CLI writes