Overview
One API for every model.
Synaps gives you Claude, GPT, Gemini, DeepSeek, Qwen and more behind a single OpenAI-compatible endpoint, billed per request from a USD balance you top up with crypto.
What you get
- One key, many models. Switch providers by changing the
modelstring. No new accounts, no separate bills. - OpenAI-compatible. Point any OpenAI SDK or tool at Synaps by changing the base URL.
- Pay per request. No subscription. Every call is billed at the provider's cost plus a flat 20% fee, and each response tells you exactly what it cost.
- Wallet account. Sign in by signing a message. Top up with ETH on Robinhood Chain or Base, or USDC on Base.
- Privacy labels. Every model says who can read your prompts.
Base URL
https://www.synaps.chat/api/v1Every endpoint in these docs is relative to this URL. Requests need an API key in the Authorization header; see Authentication.
Endpoints
| Method | Path | Purpose |
|---|---|---|
POST | /chat/completions | Generate a reply from any model. Supports streaming, tools and JSON output. Reference |
GET | /models | List every model with its context length, privacy lane and price. Reference |
A first request
curl https://www.synaps.chat/api/v1/chat/completions \
-H "Authorization: Bearer $SYNAPS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-chat",
"messages": [{"role": "user", "content": "Say hello in five words."}]
}'