⚡ QODER2API

OpenAI & Anthropic-compatible gateway for Qoder IDE accounts — single Go binary with an account pool.

Features • Quick Start • Usage • Configuration • Troubleshooting • Contact

English | Русский

---

Features

• 🚀 Single binary — HTTP gateway, account pool and embedded dashboard in one Go binary; no external services
• 🔓 OpenAI-compatible — /v1/chat/completions, /v1/models, /v1/usage; drop-in for any OpenAI-format client
• 🎭 Anthropic-compatible — /v1/messages for Claude Code and other Anthropic-format clients
• 🔄 Account pool with failover — multiple PATs, automatic switch to the next account on agent-limit errors (code 115)
• 📊 Embedded dashboard — live HTML UI at the root URL: accounts, quotas, default model, context/thinking settings — changes persist to config.json
• 🎯 Model names pass-through — clients configure real Qoder model names directly (e.g. qwen3.8-max)
• 🌐 Cross-platform — Windows, macOS, Linux
• 📦 Standalone — pure Go, minimal dependencies

💡 Works with Claude Code out of the box! Just point its env vars at the gateway and pick a Qoder model name.

---

Quick Start

1. Download

Download the latest release for your platform:

• Windows (64-bit)
• Linux (64-bit)
• macOS Intel
• macOS Apple Silicon

Or build from source:

git clone https://github.com/D3-vin/Qoder2Api.git
cd Qoder2Api
go build -trimpath -ldflags="-s -w" -o qoder2api .

2. Configure

cp .env.example .env
# Edit .env — set QODER_PAT from your Qoder account settings (see Configuration)

If .env.example is missing (bare release binary), just run the binary once — it extracts .env.example, baseprompt_min.json and baseprompt.json automatically.

3. Run

./qoder2api

Dashboard: http://127.0.0.1:8963/

4. Test

curl -X POST http://127.0.0.1:8963/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3.8-max","messages":[{"role":"user","content":"Hello!"}],"stream":false}'

---

Usage

Streaming

curl -N -X POST http://127.0.0.1:8963/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3.8-max","stream":true,"messages":[{"role":"user","content":"Say hi"}]}'

Connect Claude Code

Clients configure model names directly — point Claude Code env vars at Qoder model names:

export ANTHROPIC_BASE_URL=http://127.0.0.1:8963
export ANTHROPIC_AUTH_TOKEN=***
export ANTHROPIC_DEFAULT_HAIKU_MODEL=qwen3.8-max
export ANTHROPIC_DEFAULT_SONNET_MODEL=qwen3.8-max
export ANTHROPIC_DEFAULT_OPUS_MODEL=qwen3.8-max
export CLAUDE_CODE_SUBAGENT_MODEL=qwen3.8-max

Connect any OpenAI client

Setting	Value
Base URL	http://127.0.0.1:8963/v1
API Key	any non-empty value
Model	any name from GET /v1/models (e.g. qwen3.8-max)

Models

Live model list is served from Qoder at GET /v1/models. Built-in aliases:

Model	Qoder key
qwen3.8-max	qmodel_38max
qwen3.7-max	qmodel_latest
qwen3.7-plus	qmodel
deepseek-v4-pro	dmodel
deepseek-v4-flash	dfmodel
lite	lite
auto	auto

API endpoints

Method	Path	Description
POST	/v1/chat/completions	OpenAI chat (stream / non-stream)
POST	/v1/messages	Anthropic messages
GET	/v1/models	Live model list from Qoder
GET	/v1/usage	Active account quota
GET	/	Dashboard
GET	/api/status	Server status, active account, default model
GET	/api/accounts	List accounts with quotas
POST	/api/accounts/add	Add a PAT at runtime
POST	/api/accounts/remove	Remove an account
POST	/api/accounts/select	Switch active account
POST	/api/quota/refresh	Refresh quotas
GET	/api/promo	Promo activities and free quotas
POST	/api/settings	Default model, per-model context/thinking

---

Project Structure

├── main.go               # Entry point, .env loader, pool startup
├── server/               # HTTP server, account pool, dashboard, protocol bridges
├── api/                  # Upstream API clients (bearer, signature, OpenAPI)
├── auth/                 # Token and signature building
├── encoding/             # Qoder wire encoding
├── httputil/             # HTTP client helpers
├── assets/               # Embedded templates + .env.example (auto-extracted on first run)
├── config.json           # Runtime state (managed by dashboard)
└── README.md

---

Configuration

Settings in .env (auto-loaded) or environment variables:

Variable	Default	Description
QODER_PAT	(required)	Qoder Personal Access Token
QODER_PAT_LIST	(empty)	Multiple PATs (comma/newline separated); takes precedence over QODER_PAT, pool fails over on agent-limit errors
QODER_HOST	127.0.0.1	Bind address
QODER_PORT	8963	Server port
QODER_MODEL	lite	Default model
QODER_PROMPT_PROFILE	min	Prompt size: min (~2 KB) or full (CLI dump)
QODER_PROXY_ENABLED	false	Enable outbound proxy
QODER_PROXY_URL	(empty)	Outbound proxy URL (e.g. http://127.0.0.1:8888)
QODER_INSECURE_SKIP_VERIFY	false	Skip TLS verification for the proxy
QODER_MACHINE_ID / QODER_MACHINE_TOKEN / QODER_MACHINE_TYPE	(empty)	Machine credentials from your authenticated Qoder IDE session

Dashboard changes (PATs, active account, default model, per-model context/thinking) persist to config.json next to the binary.

---

Troubleshooting

Symptom	Cause	Fix
no PAT configured at startup	Missing token	Set QODER_PAT or QODER_PAT_LIST in .env
Agent-limit error (code 115)	Account quota exhausted	Add more PATs — the pool fails over automatically
Request fails on startup	Bad machine credentials	Set QODER_MACHINE_ID / QODER_MACHINE_TOKEN / QODER_MACHINE_TYPE from your IDE session
Full prompt missing	baseprompt.json not found	Place baseprompt.json next to the binary for QODER_PROMPT_PROFILE=full

---

Contact

• GitHub: https://github.com/D3-vin/Qoder2Api
• Telegram: @D3_vin
• Chat: @D3vin_chat
• Author: @D3vin_dev

---

⚠️ Disclaimer

This software is provided for educational purposes only, to demonstrate technical possibilities of API bridging. It is provided "as is" without any warranty. The author bears no responsibility for any consequences of its use. Users must comply with applicable laws and the terms of service of all third-party services involved.

License

MIT License — see LICENSE