Home / Live Demo

Try the models — right here

No signup needed for the demo. Pick a model and ask anything. Full access (all models, longer context, your own key) starts at the console.

What this demo runs on

This page is a live demo of the QuantoPort gateway. Instead of holding one provider account per model, you send the same OpenAI-compatible request to a single base URL (https://quantoport.com/v1) and switch models by changing the model name. Anthropic-compatible and Responses endpoints are also supported.

Billing follows our published basis: Claude and GPT, DeepSeek, Kimi and Qwen at official list rates; GLM-5.3 at 15% off. The demo is rate-limited — create an API key to use the full catalogue and your own balance.

LIVE · demo

👋 Hi! Pick a model above and ask anything — I'll respond through the QuantoPort gateway. Tip: try "write a Python function to fetch RSS" or "explain quantum computing simply".

Demo limited to 8 messages per 10 minutes per visitor. Sign up for unlimited use and all 9+ models.

About this demo

What you are chatting with, and how it differs from a paid key.

🔌 Real models, real gateway

Every message goes through the same gateway, routing and billing infrastructure as paid traffic — you are testing the production path, not a canned bot.

🎁 Free to try

No signup needed. Limited to 8 messages per 10 minutes per visitor so everyone gets a turn; the paid gateway has no such limit beyond your own key settings.

📈 When you need more

Create a key in the console and point any tool at https://quantoport.com/v1 — all 9+ models, 1M-token contexts, streaming.

🔒 About your demo messages

Demo conversations are processed like normal API traffic and are not stored by us. Do not paste secrets or personal data into any AI chat, here or elsewhere.

The demo uses non-streaming mode to keep the page simple. Paid keys support full SSE streaming, so tools like Claude Code and Cherry Studio show answers token-by-token.

Not here — the demo is deliberately minimal. With your own key you control the full request body: system prompt, temperature, tools, JSON mode, anything the upstream API supports.

Yes. A subset of the catalogue is offered in the demo to keep it fast and affordable; the paid gateway exposes the full per-key model list via GET /v1/models.