Your AI programming subscription may only be half utilized

You spend $20 a month on AI programming tools—Qoder, Codex, Claude Code—but have you considered that these subscription quotas can only be used in their own terminals? Want to use Cursor? No. Want to use LobeChat? No. Want to call the API from your own script? Even less.

Your $20 bought a VIP card that can only be used one way.

Until I installed this thing.

One command turns your subscription into an API

qorder-proxy does something simple: it translates the subscription quota from Qoder CLI into a local OpenAI / Anthropic compatible API.

After installation, your 127.0.0.1:3000 becomes a full OpenAI API endpoint. Cursor, Continue, LobeChat, ChatBox, even your own Python scripts—any client that supports the OpenAI protocol can directly call the models in your Qoder subscription.

Qwen3.7-Max, DeepSeek V4 Pro, GLM-5.2, Kimi K2.6—these top domestic models are no longer locked in the Qoder terminal. They can work in any tool you like.

This is not a relay hub. No public deployment, no multi-user sharing, no commercial operation. It runs only on your local machine, consuming only your own subscription quota.

Why do we need "translation"?

The problem lies in the protocol.

Qoder CLI has its own dialog protocol, message format, streaming response format, and tool calling format. These formats are different from OpenAI's /v1/chat/completions and also different from Anthropic's /v1/messages. It's like one person who only speaks Japanese and another who only speaks French sitting at the same table—they can see each other but can't have a conversation.

qorder-proxy is that interpreter.

It starts a local HTTP service, receives requests in OpenAI or Anthropic format, translates them into a format that Qoder CLI can understand, calls the CLI for inference, and then translates the result back into the format the client expects.

Throughout the process, your tokens never leave your machine.

3-minute setup

I tried it, and indeed it takes just 3 minutes.

# 1. Install Qoder CLI
npm install -g @qodercn-ai/qoderclicn

# 2. Clone and install the proxy
git clone https://github.com/lininn/qorder-proxy.git
cd qorder-proxy/qoder-proxy
npm install && npm link

# 3. Configure (Web interface, just fill in the Token)
qorder-proxy --web

# 4. Start
qorder-proxy start

After startup, run curl http://127.0.0.1:3000/health. If you see {"ok":true}, it's ready.

The web configuration interface is new in v2.0—previously you could only configure it via CLI command line. Now just fill in the browser, three fields: Token, backend, port. Save and it auto-closes. Ten times more comfortable than when I last configured nginx reverse proxy.

What models does it support?

The CN backend (qoder.com.cn) currently supports the following model list:

• qwen3.7-max — Tongyi Qianwen flagship, supports reasoning strength adjustment (low/medium/high/max four levels)
• deepseek-v4-pro / deepseek-v4-flash — DeepSeek V4 dual versions
• glm-5.1 / glm-5.2 — Zhipu GLM latest version
• kimi-k2.6 — Moonshot Kimi latest
• qwen3.6-plus / qwen3.6-flash — Qianwen 3.6 series

There is also an auto mode — let Qoder choose the most appropriate model itself.

What is it like using these models in Cursor? I tried Qwen3.7-Max with the Continue plugin; code completion and dialogue worked normally. DeepSeek V4 Pro for complex refactoring, speed is about the same as using the CLI directly—because it does run the CLI under the hood.

Two backends, two worlds

qorder-proxy supports dual backend switching:

One command to switch: qorder-proxy config set backend global, then qorder-proxy restart.

CN backend uses PAT authentication, created at qoder.com.cn/account/integrations, independent of any third-party login flow. Global backend uses OAuth, consistent with Qoder official.

Security boundary: it's not a relay hub

This must be clarified.

qorder-proxy listens only on 127.0.0.1 by default, does not bind to 0.0.0.0, and does not expose to the public network. Configuration file permissions are 0600, readable only by the local user. Logs automatically redact Token, Cookie, and Authorization headers.

The README states clearly: Public deployment is prohibited, multi-user sharing is prohibited, API resale is prohibited, bypassing official billing risk control is prohibited.

This is not fluff. The design logic of this tool is "personal local use"—each request starts a new CLI subprocess, not designed for concurrency at all. If you use it as a public API gateway, performance will teach you a lesson.

There are public OpenAI relay services out there—CloseAI, openai-forward, chatgptProxyAPI—they do something else: public proxy forwarding, multi-user API key sharing, billed by usage or monthly. Those are commercial services, not in the same arena as qorder-proxy.

qorder-proxy is more like a protocol adapter—you speak French, it speaks Japanese. It translates for you, and once translated, the job is done.

Limitations of tool calling

To be honest, it's not perfect.

Tool use / function calling is implemented via prompt instructions + text parsing, not native model capability. What does that mean? qorder-proxy inserts an instruction into the prompt sent to the CLI, asking the model to output tool calls in a specific format, then uses regex or JSON parsing to extract the result.

This approach has obvious drawbacks: if the model outputs non-standard JSON, or simply refuses to use the tool format, the response degrades to plain text. Additionally, requests with tools do not use streaming—because a complete JSON is needed for parsing.

But this is not a fatal flaw. Most developers use qorder-proxy for dialogue and code generation, not heavy Agent workflows. If you really need to run an Agent, using Qoder CLI directly is more appropriate.

Web Console: small but beautiful

After starting, visit http://127.0.0.1:3000/ui to access a local web console:

• Dashboard — health status, Base URL, model count
• Models — model list (calling /v1/models)
• Chat Test — simple non-streaming test
• Config — generate configuration examples for OpenAI / Anthropic / OpenCode
• Usage — local usage statistics (token level, estimated)

Usage page data is locally estimated, not a substitute for official Qoder billing. But it's sufficient for monitoring your own consumption pace—at least you can know if you've used too much today.

An interesting question

qorder-proxy was born on the edge of a gray zone.

It does not provide tokens, does not share quotas, does not bypass billing—but its existence indeed amplifies the value of a Qoder subscription. A $20 Qoder Pro, originally usable only in the CLI, now becomes a full local API ecosystem. Is this good or bad for Qoder?

Users gain more value, but perhaps that also means higher resource consumption. If everyone hooks their subscription into Cursor for intensive usage, will Qoder's compute costs become unsustainable?

Currently Qoder has not commented on this tool. The disclaimer in the README is clear—if it is not officially allowed, please stop using it immediately.

This sentence is both a boundary and a question mark hanging overhead.

Relevant links:

• qorder-proxy GitHub: https://github.com/lininn/qorder-proxy
• Qoder CN PAT creation: https://qoder.com.cn/account/integrations
• Qoder official website: https://qoder.com/zh

Free Qoder Pro: Register using my referral link to get a free month of Pro—GLM-5.2 is available for use. → https://qoder.com.cn/referral?referral_code=SOTtWioqAyhveY6SPPYSfjMIuKm4CnLz