Build with Aether AI.
A single OpenAI-compatible API endpoint for models available through your Aether AI account. Create a key, choose an available model, and send your first request.
OpenAI-compatible APIAuthentication
Every model API request needs a valid Aether AI API key. Sign in to the dashboard, open API Keys, create a key, and store it privately. Keys are account-specific; never embed a secret in public browser code.
Authorization: Bearer YOUR_AETHER_API_KEY
Quickstart
Set your Aether AI API key in an environment variable, then list the models your account can access. Model availability depends on the configured provider channels and your account's permissions.
curl "https://aether.cesiummc.space/v1/models" \ -H "Authorization: Bearer $AETHER_API_KEY"
Replace $AETHER_API_KEY with a private environment variable in your shell. In Windows PowerShell, use $env:AETHER_API_KEY instead.
List models
Call GET /v1/models to discover the model IDs currently accessible to your key. Use a returned model ID for inference requests.
/v1/modelsRequires an API key. A request without one returns HTTP 401.
Chat completions
Send messages through the OpenAI-compatible chat completions endpoint. Replace YOUR_MODEL_ID with a model reported by /v1/models.
curl "https://aether.cesiummc.space/v1/chat/completions" \
-H "Authorization: Bearer $AETHER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "YOUR_MODEL_ID",
"messages": [
{"role": "user", "content": "Hello, Aether AI!"}
]
}'Using the OpenAI SDK
Most OpenAI-compatible clients can target Aether AI by changing the base URL and supplying an Aether AI key. For example, with the Python OpenAI SDK:
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["AETHER_API_KEY"],
base_url="https://aether.cesiummc.space/v1",
)
response = client.chat.completions.create(
model="YOUR_MODEL_ID",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)Different providers and models may support different parameters. Check model capabilities before enabling advanced options.
Common errors
| HTTP | What it means | What to check |
|---|---|---|
| 401 | Missing or invalid authentication | API key and Bearer header |
| 403 | Request forbidden | Account, model, or key permissions |
| 429 | Rate or quota limit | Quota, balance, and usage limits |
| 5xx | Service or upstream error | Retry policy, upstream availability |
Quota, model availability, and pricing are controlled through the Aether AI dashboard and may change over time.
Account & security
Sign in to Aether AI to manage API keys, usage, and account settings. Treat API keys like passwords. Set expiration and quotas where available and rotate credentials if exposed. You can use enabled social sign-in providers to access your account.
Support & service status
For account issues, check the Aether AI dashboard. The API exposes /api/status for public service metadata.