API reference

The gateway implements the OpenAI REST API. Base URL: https://awsbee.com/api/v1. Authenticate with Authorization: Bearer bee_sk_…. Set model to a deployment name.

Endpoints

Method & path	Description
`GET /models`	Your deployments, with status, engine, source model, instance type and region.
`POST /chat/completions`	Chat completion. Supports streaming, tools and JSON mode when the engine does.
`POST /completions`	Text completion.
`POST /embeddings`	Embeddings, for deployments that serve an embedding model.

Streaming

Send "stream": true to receive server-sent events. The gateway adds stream_options.include_usage upstream so token counts are recorded; the final chunk carries usage.

curl

curl -N https://awsbee.com/api/v1/chat/completions \
  -H "Authorization: Bearer $BEE_API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"llama-3-1-8b-instruct","stream":true,
       "messages":[{"role":"user","content":"Count to five."}]}'

Response headers

x-bee-deployment names the deployment that served the request.

Errors

Errors use the OpenAI shape:

404

{ "error": { "message": "No deployment named \"x\". Available: …", "type": "invalid_request_error", "code": "model_not_found" } }

401	Missing, invalid or revoked API key.
404	No deployment with that name.
429	More than 600 requests per minute on one key.
502	The deployment did not answer.
503	The deployment is not running (launching, stopped, failed).