Basic
¥79 /mo
- •Daily limit: $30
- •Weekly limit: $100
- •Monthly limit: $360
- •For solo developers and light-to-mid Codex workflows
- •Includes mainstream Codex models and baseline support
Affordable for individuals · Reliable for teams
Access leading models like GPT and Claude through one provider. Individuals save on development, while companies get ongoing support from our team.
Compare prices and credit before you sign up. Already have an account? Log in to continue
Compare prices in RMB, choose subscriptions or pay-as-you-go, and track your credit and usage.
Use models for coding, chat, and business workflows without maintaining multiple integrations and bills.
From contracts and invoices to a dedicated support group and 24/7 assistance, our team helps with adoption.
Manage integrations, usage, and billing in one place. Available models and rates depend on the model list and selected plan.
Transparent pricing
Compare the same models and unit so the savings are easy to understand.
| Model | CodeProxy Input / output (CNY) | Official reference (USD) | Official in CNY | Savings (input / output) |
|---|---|---|---|---|
| Loading model prices… | ||||
Unit: per 1M tokens (input / output); reference rate: 1 USD = 7.2 CNY.
Input and output are compared separately. Savings depend on model and usage; cache, long-context and other special rates are excluded. Official reference prices are maintained in the catalog; current billing rules apply.
Start individually, scale as needed
Call multiple models flexibly or choose an AI coding subscription. New users can start small; returning users can top up or renew here.
Product-specific subscription tiers
The subscription area is now split into Codex, Claude Code, and Grok Search. Codex and Claude Code have separate pricing ladders, while Grok Search keeps its fixed package.
Built for everyday coding throughput
Codex plans focus on better price-performance and roomier limits for solo builders, small teams, and sustained engineering output.
¥79 /mo
Best-value Codex tier
¥199 /mo
¥399 /mo
Higher model cost, tighter quota design
Claude Code costs materially more than Codex, so its pricing and limits are designed separately for teams that truly need longer-context agent workflows.
¥249 /mo
Balanced Claude Code tier
¥599 /mo
¥1299 /mo
Price locked, layout refreshed only
Grok Search remains a fixed 3000-call / 180-day package. This update only changes how it is presented in the page layout.
¥50
Share useful AI, earn toward your next plan
When a friend registers through your link and completes an eligible payment, you earn points that can be redeemed for supported plans.
Benefits come from referral points; they are not granted automatically at sign-up. Rates and eligibility follow the campaign rules.
BUSINESS ACCESS / ENTERPRISE
Move from individual testing to business adoption with leading models in one place and a team to support procurement and technical questions. After confirming enterprise onboarding, arrange contracts, invoices, and a dedicated support group.
Support for procurement, finance, and reconciliation.
A dedicated team follows up after onboarding.
Reduce integrations and maintenance across providers.
Discuss usage, budget, and service plans for your needs.
Let’s discuss your business budget
Single top-up over CN¥10,000
Contact us for additional discounts; terms depend on your plan. You can discuss business access before topping up.
We can also arrange DingTalk, Feishu, or Telegram. Mention “business access” and include your use case and expected volume.
From choosing a plan to your first request
Compare model rates, plan credit, and eligibility before choosing pay-as-you-go or a coding subscription.
Confirm your order, then configure your API key and endpoint using the docs.
Contact our business team for contracts, invoices, and coordinated onboarding.
POST /v1/chat/completions
Authorization: Bearer YOUR_API_KEY
{
"model": "YOUR_MODEL",
"messages": [
{ "role": "user", "content": "Hello" }
]
}Supports compatible APIs and developer tools. See the docs for model-specific request formats and endpoints.
Currently, we support OpenAI Codex models including gpt-image-2.5,gpt-image-2,gpt-6.1-sol,gpt-6-astra,gpt-5.6-sol,gpt-5.6-luna,gpt-5.6-terra,claude-fable-5-1,claude-opus-5,claude-opus-4-6,claude-opus-4-7,claude-opus-4-8,claude-fable-5,claude-sonnet-5,claude-sonnet-4-6,claude-haiku-4-5-20251001,claude-sonnet-4-5-20250929,grok-4.3,deepseek-v4-flash,deepseek-v4-pro,kimi-k3,seedance-2.5,minimaxH3,lyria. We will continue to add more model support in the future.
Sign up and log in, create an API key in the dashboard, then use the request examples in docs.
It works with mainstream AI coding and development tools, including Codex CLI, Codex App, Codex plugins, Claude Code, OpenCode, and OpenClaw. If a tool supports compatible API keys or OpenAI/Codex-compatible integration, it can usually be configured directly.
Yes. We support both monthly subscriptions and pay-as-you-go billing, with live usage details shown in billing pages.
Check current subscriptions and available plans in the console. Renewal, switching, and quota rules follow each plan; contact support before purchasing if unsure.
Confirmed enterprise integrations can arrange contracts, invoices, and a dedicated 24/7 support group. Contact us for extra discounts on a single top-up over CNY 10,000. WeChat is preferred; DingTalk, Feishu, and Telegram are also available.
CodeProxy offers multi-model access, CNY pricing, and flexible quotas. Models, compatibility, limits, and billing follow the model catalog and chosen plan.
Resource scheduling and load balancing help continuity. Latency and availability depend on the model, request, and upstream service. Contact us to discuss enterprise integration and support.
Individuals can contact support about setup, usage, and billing. Confirmed enterprise integrations receive a dedicated service group with 24/7 support.
One place for models, billing, and support from individual development to enterprise purchasing.