Messages (Anthropic)

The POST /v1/messages endpoint is compatible with the Anthropic Messages API, so it can be used as a direct replacement for the Anthropic base URL. Point the Anthropic SDK to https://kenari.id/v1, replace the API key with a kenari key (kn-...), and responses will follow the Anthropic message shape.

Request

Required fields: model and messages. max_tokens is optional. When omitted, the server uses a default of 65,536 tokens.

Field	Type	Description
model	string	Required. Model id, for example step-3-7-flash.
max_tokens	integer	Optional. Limit for generated tokens. Defaults to 65,536 when omitted.
messages	array	Required. List of conversation messages.
system	string	Optional. System instructions.
stream	boolean	Optional. Enable streaming.
temperature	number	Optional.
top_p	number	Optional.
top_k	integer	Optional.
stop_sequences	array	Optional. List of stop strings.
tools	array	Optional. Tool definitions, see Tools.
tool_choice	object	Optional. Tool choice settings.
thinking	object	Optional. For reasoning models, see Reasoning.

messages

The messages field is an array of {role, content} objects. The role value is one of user or assistant. The content value can be a plain string, or an array of content blocks (for example text blocks or image blocks).

{
  "model": "step-3-7-flash",
  "max_tokens": 512,
  "messages": [
    {"role": "user", "content": "Halo!"}
  ]
}

Response

Responses follow the Anthropic message shape.

Field	Type	Description
id	string	Message id.
type	string	Always "message".
role	string	Always "assistant".
model	string	Model that answered.
content	array	List of content blocks, for example {"type": "text", "text": "..."}.
stop_reason	string	Stop reason.
stop_sequence	string	Matching stop sequence, or null.
usage	object	{input_tokens, output_tokens}.

{
  "id": "msg_...",
  "type": "message",
  "role": "assistant",
  "model": "step-3-7-flash",
  "content": [
    {"type": "text", "text": "Halo! Ada yang bisa saya bantu?"}
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {"input_tokens": 9, "output_tokens": 12}
}

Note Tokens in `usage` are charged from your Rupiah balance. See Billing for rate details.

Examples

Anthropic SDK Python:

import anthropic

client = anthropic.Anthropic(base_url="https://kenari.id/v1", api_key="kn-...")
message = client.messages.create(
    model="step-3-7-flash",
    max_tokens=512,
    messages=[{"role": "user", "content": "Halo!"}],
)
print(message.content[0].text)

Anthropic SDK TypeScript:

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

const client = new Anthropic({ baseURL: "https://kenari.id/v1", apiKey: "kn-..." });
const msg = await client.messages.create({
  model: "step-3-7-flash",
  max_tokens: 512,
  messages: [{ role: "user", content: "Halo!" }],
});
console.log(msg.content[0].text);

curl:

curl https://kenari.id/v1/messages \
  -H "Authorization: Bearer kn-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"step-3-7-flash","max_tokens":512,"messages":[{"role":"user","content":"Halo!"}]}'

Streaming

Set "stream": true to receive the response gradually as server-sent events. Event format details are in Streaming.