Your first API request
Create an API key in the console, then set your endpoint and model name. Elysia API is an independently operated third-party AI API relay run by an individual based in Japan.
1. Account and API key
Create an account and complete email verification and the anti-bot check. After signing in, create a new key under API keys. Use separate keys for different projects and configure their groups, allowed models and usage limits.
Treat API keys like passwords. Keep them in server-side environment variables; never publish them in browser code or Git repositories.
2. Endpoint and model
Base URL: https://api.elysia.earth/v1
Choose a model available to your API key's group from
Models & pricing. Replace
MODEL_ID below with the exact model name. Review the
model's supported endpoints. Temporarily suspended models cannot be
used.
Store your key in the ELYSIA_API_KEY environment variable
before running these examples. They require a model that supports Chat
Completions.
cURL
curl https://api.elysia.earth/v1/chat/completions \
-H "Authorization: Bearer $ELYSIA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "MODEL_ID",
"messages": [{"role": "user", "content": "Hello"}],
"max_tokens": 64
}'
Python (OpenAI-compatible client)
Install the SDK with pip install openai.
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["ELYSIA_API_KEY"],
base_url="https://api.elysia.earth/v1",
)
response = client.chat.completions.create(
model="MODEL_ID",
messages=[{"role": "user", "content": "Hello"}],
max_tokens=64,
)
print(response.choices[0].message.content)
Output-limit parameters and supported features vary by SDK and model. If you receive an unsupported-parameter error, check the model details.
3. Check prices and usage history
Input, output, cache and other billable usage are deducted from prepaid API credit. The $ symbol denotes internal service credit. The standard purchase rate is JPY 40 = $1 API credit, not a USD exchange rate or cash balance. Payments are charged in Japanese yen (JPY).
Check the pricing page for current rates and long-context pricing conditions. After a request, review token counts and charges in usage history, and balances and purchases in your wallet.
Troubleshooting
- Authentication error: Check that your key is enabled, has not expired and has sufficient quota and account balance.
- Model unavailable: Check the exact name, key group, model restrictions and whether the model is suspended.
- Request format error: Review the selected model's supported endpoint and parameters.
- Congestion or upstream error: Wait before retrying. Avoid repeated requests in a tight loop.
If the issue persists, email hello@elysia.earth with the model, time and error details. Do not send API keys, passwords or confidential prompt content. Our primary support language is Japanese.