API
Reference
Direct bridge to machine intelligence. Unified specification across 30+ LLM providers with automatic routing, failover, and observability.
Chat Completions
/v1/chat/completionsPrimary inference endpoint. Routes to the optimal provider based on your configured strategy — latency, cost, or primary/fallback.
Parameters
Model ID to use. e.g. gpt-4o, claude-3-opus, aurora-best-latency
Type: stringArray of message objects representing the conversation.
Type: arrayEnable server-sent event streaming of partial responses.
Type: booleanMaximum tokens in the generated response.
Type: integerSampling temperature 0–2. Defaults to 1.
Type: numbercurl -X POST https://api.aurora.gateway/v1/chat/completions \
-H "Authorization: Bearer $AURORA_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-4o",
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Explain quantum computing."}
],
"stream": true
}'{
"id": "aur-8f3a-x1b2",
"object": "chat.completion",
"created": 1749268800,
"model": "gpt-4o",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Quantum computing leverages qubits ..."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 28,
"completion_tokens": 142,
"total_tokens": 170
},
"provider": "openai",
"latency": "142ms"
}/v1/chat/completions?stream=trueSame as the standard endpoint but returns a Server-Sent Events stream. Each chunk contains a delta of the response.
Parameters
Model ID to use.
Type: stringConversation messages array.
Type: arrayMust be set to true.
Type: booleancurl -X POST https://api.aurora.gateway/v1/chat/completions \
-H "Authorization: Bearer $AURORA_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-3-opus",
"messages": [
{"role": "user", "content": "Write a short poem."}
],
"stream": true
}'data: {"id":"aur-9c2d","choices":[{"delta":{"role":"assistant"}}]}
data: {"id":"aur-9c2d","choices":[{"delta":{"content":"Here"}}]}
data: {"id":"aur-9c2d","choices":[{"delta":{"content":"'s a"}}]}
data: {"id":"aur-9c2d","choices":[{"delta":{"content":" poem..."}}]}
data: [DONE]Rate Limits
Requests Per Minute
Tokens Per Minute
Standard HTTP Code