API Reference

Base URLs, authentication, endpoints, request/response schemas, and error codes.

Capped AI's API is OpenAI-compatible. Any SDK or client that works with OpenAI's API works with Capped AI by changing the base URL and API key.

Base URL

Your environment has a dedicated subdomain:

https://api.{your-account}.capped.ai/v1

Replace {your-account} with the account slug provided at deployment. The exact base URL is available in your Capped AI dashboard and in AWS Secrets Manager at lomao/api-base-url.

Authentication

All requests require a bearer token in the Authorization header:

Authorization: Bearer lom_live_...

API keys are issued per application and stored in AWS Secrets Manager in your account. Rotate them any time from the Capped AI dashboard or via the Secrets Manager console.

Endpoints

Chat Completions

POST /v1/chat/completions

{
  "model": "claude-sonnet-4",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "What is the capital of France?" }
  ],
  "max_tokens": 512,
  "temperature": 0.7,
  "stream": false
}

Streaming Chat Completions

POST /v1/chat/completions

{
  "model": "claude-sonnet-4",
  "messages": [{ "role": "user", "content": "Write a haiku." }],
  "stream": true
}

// Response: SSE stream
data: {"id":"cmpl-...","choices":[{"delta":{"content":"An old silent pond"}}]}
data: {"id":"cmpl-...","choices":[{"delta":{"content":"..."}}]}
data: [DONE]

Embeddings

POST /v1/embeddings

{
  "model": "titan-embed-text-v2",
  "input": "The quick brown fox jumps over the lazy dog"
}

// Response
{
  "object": "list",
  "data": [{ "object": "embedding", "embedding": [...], "index": 0 }],
  "model": "titan-embed-text-v2",
  "usage": { "prompt_tokens": 10, "total_tokens": 10 }
}

List Models

GET /v1/models

// Response
{
  "object": "list",
  "data": [
    { "id": "claude-sonnet-4", "object": "model", "created": 1714521600 },
    { "id": "claude-haiku-3.5", "object": "model", "created": 1714521600 },
    { "id": "llama-3-70b", "object": "model", "created": 1714521600 }
  ]
}

Model Aliases

Alias	Maps to (Bedrock model ID)
claude-sonnet-4	anthropic.claude-sonnet-4-20250514-v1:0
claude-haiku-3.5	anthropic.claude-haiku-3-5-20241022-v1:0
claude-sonnet-3.5	anthropic.claude-3-5-sonnet-20241022-v2:0
llama-3-70b	meta.llama3-3-70b-instruct-v1:0
llama-3-405b	meta.llama3-1-405b-instruct-v1:0
mixtral-8x7b	mistral.mixtral-8x7b-instruct-v0:1
mistral-large	mistral.mistral-large-2402-v1:0
titan-embed-text-v2	amazon.titan-embed-text-v2:0

Error Codes

HTTP status	Code	Meaning
401	invalid_api_key	API key missing or invalid
429	rate_limit_exceeded	Usage plan quota exhausted — check your plan limits
400	invalid_model	Model alias not found in your environment
400	context_length_exceeded	Input exceeds model's context window
503	model_unavailable	Bedrock model temporarily unavailable — retry with exponential backoff
500	internal_error	Contact support@capped.ai with the request ID from the response header

Rate Limits

Rate limits are set per API key in API Gateway usage plans. Defaults vary by tier:

Tier	Requests/min	Tokens/min
Pilot	60 RPM	100K TPM
Standard	300 RPM	500K TPM
Pro	1,000 RPM	2M TPM
Enterprise	Custom	Custom

Provisioned throughput units on Pro and Enterprise tiers provide guaranteed TPM above the on-demand baseline.

Back to Docs