Integrate OpenAI Codex

Connect OpenAI Codex CLI to Puyun AI to use any large language model for AI coding.

Overview

OpenAI Codex CLI is OpenAI's official command-line AI coding tool, which natively uses the OpenAI Chat Completions API protocol. With Puyun AI, you can call Claude, Gemini, and other models in Codex.

Configuration

Method 1: Environment Variables

Set the following environment variables in your terminal:

export OPENAI_API_KEY="sk-your-api-key"
export OPENAI_BASE_URL="https://ai.tracup.com/v1"

Then run Codex:

codex

Method 2: Configuration File

Edit the Codex configuration file ~/.codex/config.json:

{
  "provider": "openai",
  "apiKey": "sk-your-api-key",
  "baseURL": "https://ai.tracup.com/v1",
  "model": "gpt-4o"
}

Using Different Models

Codex supports switching models via parameters or configuration:

# Use GPT-4o
codex --model gpt-4o

# Use Claude Sonnet (automatic protocol conversion)
codex --model claude-sonnet-4-6

# Use Gemini 2.5 Pro (automatic protocol conversion)
codex --model gemini-2.5-pro

# Use o3 reasoning model
codex --model o3

Automatic Protocol Conversion

Codex sends requests using the OpenAI protocol. When you call non-OpenAI models (such as Claude or Gemini) through Puyun AI, the gateway automatically:

Converts OpenAI-format requests to the target model's native format
Converts the target model's native responses to OpenAI format
Automatically adapts SSE event format for streaming responses

Verify Connection

codex --model gpt-4o "Hello, verifying connection"

If you receive a normal response, the connection is configured successfully.

Recommended Models

Use Case	Recommended Model	Notes
General coding	claude-sonnet-4-6	Excellent code understanding and generation
Fast completion	gpt-4o-mini	Fast response, low cost
Deep reasoning	o3	Suitable for complex logic and architecture design
Best value	deepseek-chat	Good performance, low price

Troubleshooting

OPENAI_BASE_URL Not Taking Effect

Make sure the URL includes the /v1 suffix: https://ai.tracup.com/v1. Codex appends endpoint paths (e.g., /chat/completions) to the base URL.

Invalid Model Name

Codex may validate model names. If you encounter an unrecognized model name error, try using a standard model name.

Interrupted Responses

If responses are interrupted during long code generation, it may be due to max_tokens limits or service tier rate limits. Try using the Enterprise tier or reducing request complexity.