We route to the world's cheapest AI labs through an OpenAI-compatible endpoint. Prices are ~25% above upstream — still 60–80% below OpenAI and Anthropic. All billing in euros with VAT invoices included.
One account. One API key. One EUR invoice.
Proper VAT invoices on every charge. Monthly accounting standard — no currency conversion, no USD surprises on your ledger.
See EUR prices →Data Processing Agreement with sub-processor list included. Ready for your legal team — no extra procurement cycles.
View DPA →OpenAI-compatible endpoint for DeepSeek, ByteDance Doubao, and Alibaba Qwen. Change the base URL — your code stays the same.
Read the docs →Create a free account. We generate an API key instantly — no credit card required.
Top up from €50 via Stripe. EUR invoice issued automatically. Credits never expire.
Point your SDK at https://tokenbridge-zoos.polsia.app/v1. Use your API key. Done.
No subscription. No minimums. Your wallet is debited per request at TokenBridge rates.
curl https://tokenbridge-zoos.polsia.app/v1/chat/completions \
-H "Authorization: Bearer tb_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-chat", "messages": [{"role": "user", "content": "Hello, world!"}]}'API requests are processed by our upstream providers (DeepSeek in China, ByteDance in China, Alibaba in China). Your prompts and completions pass through their infrastructure. We do not store API call contents. Our full sub-processor list and data residency documentation are available at /legal/data-residency.
TokenBridge adds approximately 25% above the upstream provider's USD pricing. USD costs are converted to EUR at a 0.92 exchange rate. For example: DeepSeek-Chat input costs ~$0.14/M tokens upstream → ~€0.18/M through TokenBridge. Full EUR pricing per model is on the /models page.
Credits purchased via Stripe are non-refundable once the wallet is credited (upon checkout.session.completed webhook). For billing disputes, contact support@tokenbridge.io. We handle cases of double-charging or technical errors within 5 business days.
Yes. TokenBridge proxies streaming responses from all upstream providers. Send stream: true in your request body — responses are returned as Server-Sent Events in OpenAI-compatible format. Works with all major SDKs.
Rate limits are set per upstream provider. DeepSeek allows higher throughput than most competitors. If you need dedicated throughput (for high-volume customers), contact us before scaling. Current soft limits: 500 req/min per API key. Hard limits are configurable on request.
No free trial — we bill at cost-plus for a sustainable service. You can create a free account to see your API key and explore the dashboard. Add €50 to start making live calls. The /docs page works without a key so you can evaluate the integration before committing.
The DPA covers the data processing activities performed by TokenBridge and its sub-processors. It includes: scope of processing, data categories, retention periods, security measures, sub-processor list, and your rights as a data controller. Download it at /legal/dpa.
Same model quality. Same SDK. 60–80% less. EUR invoiced with GDPR documentation included.