Codex
OpenAI's Open-Source Coding Agent
OpenAI's open-source local coding agent that reads your codebase, edits across files, runs commands, and debugs issues — right in your terminal and IDE. Through the Ace Data Cloud OpenAI Responses-compatible proxy, enjoy lower prices and ample quota.
What is Codex?
Codex is an open-source local coding agent from OpenAI that runs in your terminal. It's not a simple autocomplete tool but a true coding agent — just describe what you want in natural language, and Codex understands your entire codebase, edits code across files, runs commands in a sandbox, debugs problems, and iterates until the task is done.
Ace Data Cloud provides a low-cost OpenAI Responses-compatible proxy, letting you use Codex at under half the official price. No fixed monthly subscription, pay per token, ample quota, globally available, and ready to use after a one-time setup.
Core Capabilities of Codex
More than autocomplete — an agent that understands, edits, executes and ships code
Codebase Understanding
Automatically analyzes project structure and dependencies — no need to manually pick context files. Just say "explain this project structure" for a full architectural overview.
Multi-file Editing
Makes precise edits across multiple files instead of just giving suggestions. Codex understands the relationships between files to keep changes correct and consistent.
Debug & Fix
Describe a bug or paste an error message, and Codex locates the root cause and implements a fix. It reads logs, runs tests, and iterates until the problem is solved.
Command Execution
Runs shell commands, executes tests, installs dependencies, and inspects Git state in a controlled sandbox — letting the agent truly complete end-to-end development tasks.
MCP Tool Integration
Supports the Model Context Protocol to connect AceData Cloud MCP tools such as Suno, Midjourney, and SERP — extending what Codex can do.
Adjustable Reasoning
Switch model_reasoning_effort between low / medium / high — let the model think deeply on complex architectural decisions and deliver well-considered solutions.
model_provider = "acedatacloud" model = "gpt-5" [model_providers.acedatacloud] name = "Ace Data Cloud" base_url = "https://api.acedata.cloud/v1" env_key = "ACEDATACLOUD_API_KEY" wire_api = "responses"
Configure in 3 Minutes
Just declare a custom provider in your Codex config to connect Codex to Ace Data Cloud.
Get an API Token
Register on Ace Data Cloud, get a Bearer Token, and set it as the ACEDATACLOUD_API_KEY environment variable
Configure config.toml
Declare a model_provider in ~/.codex/config.toml pointing to the Ace Data Cloud proxy
Start Coding
Run the codex command and enjoy full functionality at a lower cost
Multi-platform Support
Use Codex in the development environment you're comfortable with
Terminal (CLI)
The core experience — run codex in any terminal to get started
VS Code
Official extension — chat in the sidebar, reference files, preview diffs
Setup Overview
Pick the right guide for your scenario and get connected fast
MCP Tools
Connect the AceData Cloud MCP tool suite to Codex
What Can Codex Do?
From everyday coding to large-scale refactoring — covering the whole development lifecycle
Feature Development
Describe requirements in natural language, and Codex generates the complete implementation
Bug Fixing
Paste an error log or describe the issue — Codex locates the root cause and fixes it
Code Review
Reviews code quality, security risks, and potential bugs, with improvement suggestions
Refactoring
Large-scale renames, module extraction, framework migration — refactor safely across files
Get Started in 3 Steps
From sign-up to using Codex in under 5 minutes
Register & Get a Token
Create a free Ace Data Cloud account. Go to the OpenAI service page, click Acquire to get a Bearer Token, and receive free quota on first sign-up.
Configure config.toml
Declare a model_provider in ~/.codex/config.toml, set base_url and wire_api = "responses", pointing Codex to the Ace Data Cloud proxy.
Start Coding
Run codex in the terminal, or use the official extension in VS Code. Enjoy the full Codex experience at a lower cost.
Why Choose AceData Cloud?
Compared to official subscriptions, AceData Cloud wins on price, quota, and flexibility
| Comparison | AceData Cloud | Official Subscription |
|---|---|---|
| Billing | Pay per token | Fixed $20-$200 / month |
| Usage Limits | ✓ Ample concurrent quota | Strict limits, throttled at high frequency |
| Entry Price | Under 50% of official | From $20 / month |
| Over-quota Throttling | ✓ No throttling | ✗ Stops when used up |
| Global Availability | ✓ | ✗ Restricted in some regions |
| Sign-up Barrier | ✓ Ready instantly | ✗ Requires a credit card |
Codex Pricing
Transparent pay-as-you-go pricing. No subscription. No hidden fees. Pay only for the tokens you use.
Through the OpenAI Responses-compatible proxy, billed by actual token usage at under half the official price
- ✓ GPT-5: flagship coding & reasoning
- ✓ GPT-5 mini / nano: low-cost, high-throughput
- ✓ Supports the OpenAI Responses API
- ✓ Adjustable reasoning (low/medium/high)
- ✓ Free quota on first sign-up
- ✓ No monthly fee, no subscription
Dedicated plans for high-volume teams
- ✓ Volume-based bulk discounts
- ✓ Priority technical support
- ✓ Dedicated account manager
- ✓ Custom rate limits
- ✓ SLA guarantee
Frequently Asked Questions
Everything you need to know about using Codex through AceData Cloud
What's the relationship between Codex and the OpenAI API? ▾
Codex is OpenAI's open-source command-line coding agent that communicates with the models via the OpenAI Responses API under the hood. AceData Cloud provides an OpenAI Responses-compatible proxy, and Codex uses our service by configuring a custom model_provider in ~/.codex/config.toml — functionally identical to the official service.
What are the advantages of using AceData Cloud? ▾
Three key advantages: (1) Lower price — billed by token usage at under half the official price, with no fixed monthly fee; (2) Ample quota — no subscription limit prompts, no throttling at high frequency; (3) Global availability — stable, low-latency access from any region.
Which models are supported? ▾
The full GPT-5 lineup is supported, including GPT-5, GPT-5 mini, and GPT-5 nano. Set your default model in the model field of ~/.codex/config.toml, or switch it with the /model command in the Codex UI. We add new OpenAI models as soon as they're released.
How do I configure Codex to use AceData Cloud? ▾
In ~/.codex/config.toml, declare model_provider = "acedatacloud", and under [model_providers.acedatacloud] set base_url = "https://api.acedata.cloud/v1", env_key = "ACEDATACLOUD_API_KEY", and wire_api = "responses". See the setup guide for detailed steps.
Which platforms are supported? ▾
Codex supports the terminal (CLI) and the official VS Code extension, and is also compatible with VS Code-based editors like Cursor. All platforms can use the AceData Cloud service by configuring a custom model_provider.
How do I verify the configuration works? ▾
After configuring, run codex to enter the interactive UI and type /model to check that the current model and provider come from acedatacloud. You can also review API call records and usage stats in the Ace Data Cloud console to confirm requests are routed through the proxy.
Is it available worldwide? ▾
Yes. AceData Cloud provides globally available access nodes, so you get stable, low-latency Codex access from any region. No VPN, no overseas credit card — just sign up and go.
Other AI Services
Explore our full AI API suite spanning chat, image, video, music and more
OpenAI API
OpenAI GPT full model lineup, supporting Chat Completions and Responses formats
Claude API
Anthropic Claude full model lineup, supporting Chat Completion and Messages formats
Gemini API
Google Gemini full model lineup, with multimodal and long-context support
Midjourney API
World-class image generation model, integrated via REST API
Get Started with Codex Now
Through the AceData Cloud proxy, enjoy OpenAI's open-source coding agent at a lower cost. Pay as you go, no subscription, configured in 3 minutes.