Simple, Transparent Pricing
Direct USD billing, token prices match official rates: Input $1.25/M tokens, Cached input $0.125/M tokens, Output $10/M tokens
Usage & Cost Calculation
Better understand token consumption and cost calculation
Pricing Table
Standard Conversation Example
No cache hit, charged at standard rates
Cached Conversation Example
Most input hits cache, saving 90% on input costs
Why are our prices so affordable?
Our system averages only $0.02 per conversation, with pricing based entirely on official token rates. Thanks to extensive input token caching, the actual cost per conversation is remarkably low. This means you can enjoy the same AI capabilities as the official service at a fraction of the cost.
Supported GPT Models
Comprehensive support for OpenAI's latest model series, meeting diverse development needs
Basic Series
GPT-5
Latest base model with visual editor support
GPT-5 High
High-performance version for complex tasks
GPT-5 Medium
Balanced version, optimal choice for performance and cost
GPT-5 Low
Economic version for lightweight tasks
GPT-5 Minimal
Lightweight version with fast response
Codex Series
GPT-5 Codex
Programming-optimized version for code generation
GPT-5 Codex High
Codex high-performance for complex code generation
GPT-5 Codex Medium
Codex balanced version, preferred for code optimization
GPT-5 Codex Low
Codex economic version for lightweight code tasks
FAQ
Answers to your common questions
What payment methods are available?
We support Alipay for subscription plans and Xianyu for activation codes. Both methods ensure secure and convenient payments.
How to check balance and usage details?
Login to your console to view real-time balance, token usage details, and cost statistics. The system provides detailed usage records for complete transparency.
Do you provide invoices?
Yes, we provide VAT normal and special invoices. Contact support after purchase to apply; issuance takes 3–5 business days.
What happens when quota runs out?
When quota is low, you can directly renew your subscription to continue using. The system notifies you before balance runs low to ensure uninterrupted service.
How much does a conversation cost?
Conversations typically use tens to hundreds of thousands of tokens, costing around $0.02-0.03. Due to extensive caching, hundreds of thousands of tokens usually cost under $0.05. Many scenarios cost only $0.002 (full cache). You can view detailed token usage for each conversation in the console.
Which programming languages are supported?
CodexZH supports 50+ mainstream languages (Python, JavaScript, Java, C++, Go, Rust, TypeScript, and more), updated continuously.
How do you ensure data security?
We use enterprise-grade measures: end-to-end encryption, ISO 27001, and regular audits. Your code is only used for generation and never stored or reused.
Still have questions?
Our support team is here to help