Messages

curl https://api.ailoot.ai/v1/messages \
  -H "x-api-key: $AILOOT_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello! What can you do?"}]
  }' 

import anthropic
import os

client = anthropic.Anthropic(max_retries=0, 
    base_url="https://api.ailoot.ai",   # without /v1: the SDK adds /v1/messages itself
    api_key=os.environ["AILOOT_API_KEY"],
)

msg = client.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Explain what an API is in two sentences."}],
)
print(msg.content[0].text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({ maxRetries: 0,
  baseURL: "https://api.ailoot.ai", // without /v1
  apiKey: process.env.AILOOT_API_KEY,
});

const msg = await client.messages.create({
  model: "claude-sonnet-5-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Explain what an API is in two sentences." }],
});
console.log(msg.content[0].type === "text" ? msg.content[0].text : msg.content);

{
  "id": "msg_01H…",
  "type": "message",
  "role": "assistant",
  "model": "claude-sonnet-5-5",
  "content": [
    { "type": "text", "text": "Hello! I can write and explain code, answer questions…" }
  ],
  "stop_reason": "end_turn",
  "usage": { "input_tokens": 14, "output_tokens": 52 }
}

Authentication

• x-api-keyheaderrequired

  Key sk-…. You can also use Authorization: Bearer sk-… — this is what Claude Code uses with ANTHROPIC_AUTH_TOKEN.

• anthropic-versionheaderoptional

  API version, for example 2023-06-01. SDKs send it automatically.

Request body

• modelstringrequired

  Model ID that supports the Messages format — 776 IDs: Claude, as well as GPT, Gemini, DeepSeek, Qwen, and others.

• messagesarrayrequired

  Messages user and assistant; content — a string or blocks (text, image, tool_use, tool_result).

• max_tokensintegerrequired

  Maximum number of response tokens. Included in the reserve; if your balance is too low for the full limit, Ailoot reduces it, but never below 256 (1025 with thinking enabled), and reports the new limit in the header X-Ailoot-Max-Tokens-Capped.

• systemstring | arrayoptional

  System instruction.

• streambooleanoptional

  SSE event stream (message_start, content_block_delta, …, message_stop). Billable text is delivered in a single block after the cost is calculated.

• temperaturenumberoptional

  Response randomness, from 0 to 1.

• toolsarrayoptional

  Tools: name, description, input_schema.

• tool_choiceobjectoptional

  auto, any, tool or none.

• thinkingobjectoptional

  {"type": "enabled", "budget_tokens": N} — thinking for models that support it.

• stop_sequencesarrayoptional

  Strings that stop generation.

Response

• contentarray

  Response blocks: text, thinking, tool_use.

• stop_reasonstring

  end_turn, max_tokens, stop_sequence or tool_use.

• usageobject

  input_tokens, output_tokens, for caching — cache_read_input_tokens and cache_creation_input_tokens.