Every frontier model,
one magic key.
GPT, Claude, Gemini, Grok, DeepSeek and Qwen answer to the same endpoint. See every price up front, top up when you need, and let routing do the trick.
Latest flagships
The newest releases from each lab, at official list prices
OpenAI's top publicly available API model, with reasoning support and up to 128K output tokens.
Anthropic model for long-running agentic coding and knowledge work, with adaptive thinking always on.
Google's most capable Gemini API model, with multimodal input and a 1M-token context window.
xAI's flagship for coding, agentic tool calling and knowledge work, with configurable reasoning effort.
DeepSeek's newest model, with native multimodal input and thinking and non-thinking modes.
Alibaba's 2.4T-parameter MoE flagship for long-horizon coding and professional tasks.
Point your existing SDK at magicai.lol/v1 and change the model name. No new client library to learn.
Each request goes straight to the upstream provider that serves the model, with retries handled for you.
Per-token rates are listed next to every model. Add credit when you need it and spend it on any model.
curl https://magicai.lol/v1/chat/completions \
-H "Authorization: Bearer $MAGICAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-6-luna","messages":[{"role":"user","content":"Hello!"}]}'