Bridge by Wekoya / Guides / Use Codex CLI with Bridge
Use Codex CLI with Bridge
In short: Add Bridge as a model provider in ~/.codex/config.toml with wire_api = "responses", put your Bridge key in WEKOYA_API_KEY, and run codex. Codex then works with Claude, Gemini, GPT or any Bridge model that supports tool calling, billed in naira from your balance.
Updated
Set it up
- Create a free account with your email.
- Open Billing in the console and top up your naira balance by bank transfer to your own account number.
- Create a key under API keys. It starts with
wk_live_and is shown once, so keep it somewhere safe. - Add this to
~/.codex/config.toml(Codex ignores providers in a project's own config file):
model = "claude-sonnet-5-5"
model_provider = "bridge"
[model_providers.bridge]
name = "Bridge by Wekoya"
base_url = "https://wekoya-api.wekoya.tech/v1"
env_key = "WEKOYA_API_KEY"
wire_api = "responses"export WEKOYA_API_KEY=wk_live_...
codexWhat to know
- Codex only speaks OpenAI's Responses API (its own config reference says so), which Bridge serves at
/v1/responses. See Responses API and Codex. - Pick a model with tool calling ("Tool calling: Yes" on its page); Codex edits files and runs commands through tools.
- Codex's built-in web search is not available through Bridge; Codex works without it.
- Change models by editing
model, or for one run withcodex -m gemini-3.5-flash-lite.
Common questions
Can Codex use Claude models?
Yes, through Bridge. Set model to a Claude ID such as claude-sonnet-5-5 with Bridge as the provider.
Why does Codex need the Responses API?
Current Codex releases only accept wire_api = "responses" for custom providers. Bridge answers the Responses API, so Codex works with it.