camelAI Documentation

API Reference

API Endpoints

One API key works across the OpenAI and Anthropic protocols

MethodPathFormatUse
POST/v1/chat/completionsOpenAIChat and tool-use completions
POST/v1/responsesOpenAIResponses API input and streaming
POST/v1/messagesAnthropicMessages API input and streaming
POST/v1/messages/count_tokensAnthropicCount tokens before sending
GET/v1/modelsBothList available models

Chat Completions

POST /v1/chat/completions

Use the familiar OpenAI chat schema. Send a messages array and set stream to true for SSE output. System messages, tool definitions, tool choice, temperature, and output limits are passed through.

cURL
curl --no-buffer https://stream.camelai.com/v1/chat/completions \
  -H "Authorization: Bearer $CAMEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [
      {"role": "system", "content": "You are a concise coding assistant."},
      {"role": "user", "content": "Explain this error message."}
    ],
    "stream": true
  }'

Responses

POST /v1/responses

Use the OpenAI Responses schema for string or structured array input. This is the preferred endpoint for clients such as Codex that speak the Responses protocol.

cURL
curl --no-buffer https://stream.camelai.com/v1/responses \
  -H "Authorization: Bearer $CAMEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "input": "Give me three names for a database migration tool.",
    "stream": true
  }'

Messages

POST /v1/messages

Use the Anthropic Messages schema. Include max_tokens and an anthropic-version header. Set stream to true for Anthropic-format SSE events.

cURL
curl --no-buffer https://stream.camelai.com/v1/messages \
  -H "x-api-key: $CAMEL_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "How should I structure a TypeScript worker?"}
    ],
    "stream": true
  }'

Count message tokens

POST /v1/messages/count_tokens

Estimate the input-token count for an Anthropic Messages request before generating a response. Token counting also uses a stream slot and may wait in the account queue.

cURL
curl https://stream.camelai.com/v1/messages/count_tokens \
  -H "x-api-key: $CAMEL_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [
      {"role": "user", "content": "How many tokens are in this request?"}
    ]
  }'

List models

GET /v1/models

Return the model IDs currently accepted by camelStream. The response is compatible with OpenAI model discovery and also includes fields used by Anthropic gateway discovery. The models serving those IDs are described on the fleet page.

cURL
curl https://stream.camelai.com/v1/models \
  -H "Authorization: Bearer $CAMEL_API_KEY"
PreviousAuthenticationNext Management API