Pay-as-you-go gateway. No SMS verification. Instant API key. Drop-in compatible with Claude Code, Codex CLI, Cursor, Cline, and major SDKs.
One key routes Claude, GPT / Codex, and leading coding models through native protocols — with usage you can watch and billing you can predict.
No subscription, no minimums. Top up any amount, use what you need.
Get your API key right after payment. No KYC, no waiting, no SMS.
Anthropic Messages + OpenAI Responses & Chat Completions — single key, smart routing.
Check spend and quota any time on the stats page with your own key.
Works with Claude Code, Codex CLI, Cursor, Cline, Continue, LangChain, and OpenAI/Anthropic SDKs.
Full SSE streaming and tool-use support — same wire format as official APIs.
From zero to your first request in a couple of minutes.
Top up and receive an API key instantly — no identity check, no waiting.
Change the base URL and paste your key. Your existing SDK or CLI keeps working.
Call Claude, GPT-6.2, GLM, Kimi, DeepSeek, and Qwen through one key. Watch spend live on the stats page.
No subscription and no minimum — each request is metered by actual token usage against your prepaid balance.
No monthly fee and no minimum spend. Top up any amount and draw it down as you go.
Each request is metered by the tokens it actually uses; unit prices vary by model.
Claude, GPT / Codex, GLM, Kimi, DeepSeek, and Qwen billed together against a single balance.
Every request is metered in real time. Check remaining quota any moment on the stats page.
Need volume rates or a custom plan? Talk to us.
Compatible with the official wire formats — point your existing SDK at the URL below and use your key.
Set two environment variables, run your existing tool. That's it.
# ~/.bashrc export ANTHROPIC_BASE_URL="https://r-8g8.qcode.cc/api" export ANTHROPIC_AUTH_TOKEN="cr_..." claude
# ~/.codex/config.toml [model_providers.provider] name="provider" base_url="https://r-8g8.qcode.cc/openai" env_key="OPENAI_API_KEY"
# .env OPENAI_BASE_URL="https://r-8g8.qcode.cc/openai/v1" OPENAI_API_KEY="cr_..."
Already using another client? Point its base URL at the gateway and paste your key.
Latest production models across vendors — keep your client up to date and new models become available automatically.
| Model | Context | Best for |
|---|---|---|
| claude-fable-5-2 anthropic | 1M | flagship · agent |
| claude-opus-5-5 anthropic | 1M | flagship · agent |
| claude-sonnet-5-5 anthropic | 1M | coding · long context |
| claude-haiku-5-5 anthropic | 1M | fast · high throughput |
| claude-fable-5 anthropic | 1M | flagship · agent |
| claude-opus-5 anthropic | 1M | flagship · agent |
| claude-sonnet-5 anthropic | 1M | coding · long context |
| gpt-6.2-astra openai | 1.05M | flagship · reasoning |
| gpt-6.2-sol openai | 1.05M | flagship · reasoning |
| gpt-6.1-sol openai | 1.05M | flagship · reasoning |
| gpt-6-sol openai | 1.05M | flagship · reasoning |
| glm-5.3 zhipu | 200K | coding · multi-agent |
| kimi-k3.1-pro moonshot | 256K | coding · multi-agent |
| deepseek-v4.1-flash deepseek | 128K | fast · high throughput |
| qwen3.8-max qwen | 256K | coding · multi-agent |
Enter your key on the stats page to see today's spend, remaining quota, and rate limits — updated live.
One key for Claude, GPT-6.2, and frontier coding models — with usage you can watch in real time.