API Docs
https://17astra.com/v1
Quick start
Sign in to your Astra-api account, check your balance, and create a limited-budget API token in the Console. Copy the exact model ID from the Models page.
Base URL: https://17astra.com/v1. Use Authorization: Bearer <your token>. The model IDs in the code samples are placeholders, not promises of model availability.
Protocols and streaming
Chat models use the endpoint listed for that model, such as /v1/chat/completions, /v1/responses or /v1/messages. Audio, images, video and embeddings require their own documented endpoints; selecting them in a text-only chat interface is not a valid test.
For SSE, read the entire stream and check its completion event and error events. An HTTP 200 or a first text fragment alone does not prove successful completion. Do not automatically retry a potentially billed request.
Errors and billing
For 401, check token validity; for 403, check permissions and the model group; for 429, reduce concurrency and request rate. For balance errors, check both account balance and the token budget. Report 5xx errors with the timestamp, endpoint, model and request ID.
Prices and usage records are shown in the Console. Online payments are not enabled in this launch phase; contact your site administrator about quota. Never send a password or complete token to support.
curl -X POST "https://17astra.com/v1/chat/completions" \
-H "Authorization: Bearer $ASTRA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "model-name",
"messages": [
{ "role": "user", "content": "Hello" }
]
}'{
"choices": [
{ "message": { "role": "assistant", "content": "..." } }
],
"usage": { "total_tokens": 15 }
}Example model IDs are placeholders. Replace them with actual catalog IDs; availability and pricing are shown in the catalog and console.