Quick AI Coding

Developer tool guide

Connect Codex CLI to Quick AI Coding

Create a separate custom provider in your user-level Codex configuration and keep the platform key in an environment variable.

Open the verified example

1. Discover an available model

Export QUICKAICODE_API_KEY, call GET https://quickaicode.com/v1/models, and copy a returned model ID. Do not assume that every OpenAI model name is available.

2. Configure the custom provider

Merge this block into ~/.codex/config.toml and replace the model placeholder:

~/.codex/config.toml
model = "returned-model-id"
model_provider = "quickaicode"

[model_providers.quickaicode]
name = "Quick AI Coding"
base_url = "https://quickaicode.com/v1"
env_key = "QUICKAICODE_API_KEY"
wire_api = "responses"

The provider ID must not be openai, ollama, or lmstudio, which Codex reserves.

3. Export the key

Shell
export QUICKAICODE_API_KEY="your-platform-api-key"
codex

Do not paste this third-party platform key into an OpenAI account login flow or commit it to TOML.

Compatibility limits

Custom provider behavior depends on the selected model

Codex features can require specific Responses API fields or events. Test editing, tool calls, and long streaming sessions before relying on a model in production work.

Quick AI Coding is independently operated and is not affiliated with or endorsed by OpenAI.