ewpireAPI
One API. Every text model.
Call 33 text models from 11 labs with one key and one credit balance. The API speaks the OpenAI format, so the official SDKs work once you point them at ewpire.
API keys open soon. The docs are ready now, so you can plan your integration.
Base URLhttps://ewpire.com/api/v1
curl https://ewpire.com/api/v1/chat/completions \
-H "Authorization: Bearer $EWPIRE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"auto","messages":[{"role":"user","content":"Hello!"}]}'Built for the code you already have
OpenAI-compatible
The same request and answer format. Change the base URL and the key; the rest of your code stays.
Every text model, one key
Text models from many labs behind one key. Switching models is changing one string.
Auto
Pass "auto" and ewpire picks a model for each request, and moves to the next one if it is unavailable.
One balance
The API and the workspace spend the same credits. Every answer tells you what it cost and what is left.
Limits per key
Requests per minute and daily or monthly spend limits for each key. Revoke a key the moment it leaks.
Safe retries
Send an Idempotency-Key and a repeated request gets the first answer back, never a second charge.
Pick a model, or let Auto pick
| Model | Input / 1M tokens | Output / 1M tokens |
|---|---|---|
| Claude Opus 5.5 | 4.22 | 21.1 |
| GPT-6 Sol | 2.11 | 10.55 |
| Claude Sonnet 5 | 2.11 | 10.55 |
| Gemini 3.8 Flash | 0.7913 | 3.9563 |
| GPT-6 Luna | 0.1055 | 0.5275 |
| DeepSeek V4.1 Flash | 0.1583 | 0.633 |
| DeepSeek V4 Pro 0423 | 0.9004 | 1.8008 |
| Claude Opus 5 | 5.275 | 26.375 |
Pay for what you call
No subscription and no monthly minimum. Add from 10 to 2,000 credits at a time, plus a 5% service fee. Each request is charged at the model's listed price, and credits last 12 months from the day you add them.
See pricingImages, video, music and consensus
Through the API they are coming soon. In the ewpire workspace they work today.
Plan your integration now.
The docs are published and the requests follow the OpenAI format. API keys open soon; until then, every model works in the ewpire workspace.