Chat and streaming

POST/v1/chat/completions

The main endpoint for text models. Its request and response format match OpenAI Chat Completions.

Request

curl https://api-dev.satulink.ai/v1/chat/completions \
  -H "Authorization: Bearer $SATULINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v-4.1-flash",
    "messages": [{ "role": "user", "content": "Halo!" }],
    "max_tokens": 512
  }'

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.SATULINK_API_KEY,
  baseURL: "https://api-dev.satulink.ai/v1",
});

const completion = await client.chat.completions.create({
  model: "deepseek-v-4.1-flash",
  messages: [{ role: "user", content: "Halo!" }],
  max_tokens: 512,
});

console.log(completion.choices[0].message.content);

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["SATULINK_API_KEY"],
    base_url="https://api-dev.satulink.ai/v1",
)

completion = client.chat.completions.create(
    model="deepseek-v-4.1-flash",
    messages=[{"role": "user", "content": "Halo!"}],
    max_tokens=512,
)

print(completion.choices[0].message.content)

<?php
$ch = curl_init("https://api-dev.satulink.ai/v1/chat/completions");

curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer " . getenv("SATULINK_API_KEY"),
        "Content-Type: application/json",
    ],
    CURLOPT_POSTFIELDS => json_encode([
        "model" => "deepseek-v-4.1-flash",
        "messages" => [["role" => "user", "content" => "Halo!"]],
        "max_tokens" => 512,
    ]),
]);

echo curl_exec($ch);

Request fields

Field	Type	Description
`model`	string	Required. The public model ID.
`messages`	array	Required. A list of messages with a role and content.
`stream`	boolean	Optional. `true` for an event-stream response.
`max_tokens`	number	Optional but recommended. Without it the provider uses the full context and a large amount of balance is reserved.
`temperature`	number	Optional. Passed through to the provider.
`top_p`	number	Optional. Passed through to the provider.
`stop`	string / array	Optional. Passed through to the provider.
`tools`	array	Optional. See Tool calling.
`tool_choice`	string / object	Optional. Passed through to the provider.
`response_format`	object	Optional. Passed through as-is.

Non-streaming response

200 OK

json

{
  "id": "chatcmpl-7c1a",
  "object": "chat.completion",
  "created": 1760000000,
  "model": "deepseek-v-4.1-flash",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Halo! Ada yang bisa dibantu?" },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 9,
    "total_tokens": 21,
    "prompt_tokens_details": { "cached_tokens": 0 },
    "cost_idr": 38
  }
}

finish_reason holds stop, length, tool_calls, function_call, or content_filter.

Streaming

POST/v1/chat/completions

With "stream": true, the response is text/event-stream. Each line is data: <json> and entries are separated by one blank line. There is no event: name, and the stream always closes with data: [DONE].

Chunk order

text/event-stream

bash

data: {"id":"chatcmpl-7c1a","object":"chat.completion.chunk","created":1760000000,"model":"deepseek-v-4.1-flash","choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}

data: {"id":"chatcmpl-7c1a","object":"chat.completion.chunk","created":1760000000,"model":"deepseek-v-4.1-flash","choices":[{"index":0,"delta":{"content":"Ha"},"finish_reason":null}]}

data: {"id":"chatcmpl-7c1a","object":"chat.completion.chunk","created":1760000000,"model":"deepseek-v-4.1-flash","choices":[{"index":0,"delta":{"content":"lo!"},"finish_reason":null}]}

data: {"id":"chatcmpl-7c1a","object":"chat.completion.chunk","created":1760000000,"model":"deepseek-v-4.1-flash","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: {"id":"chatcmpl-7c1a","object":"chat.completion.chunk","created":1760000000,"model":"deepseek-v-4.1-flash","choices":[],"usage":{"prompt_tokens":12,"completion_tokens":9,"total_tokens":21,"cost_idr":38}}

data: [DONE]

A mid-stream failure arrives as data: {"error":{…,"code":"stream_error"}} followed by [DONE]. An error before the stream starts is still plain JSON with a normal HTTP status.

Reading a stream with the SDK

Request

curl -N https://api-dev.satulink.ai/v1/chat/completions \
  -H "Authorization: Bearer $SATULINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v-4.1-flash",
    "messages": [{ "role": "user", "content": "Halo!" }],
    "stream": true
  }'

// client dibuat seperti di bagian SDK
const stream = await client.chat.completions.create({
  model: "deepseek-v-4.1-flash",
  messages: [{ role: "user", content: "Halo!" }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

# client dibuat seperti di bagian SDK
stream = client.chat.completions.create(
    model="deepseek-v-4.1-flash",
    messages=[{"role": "user", "content": "Halo!"}],
    stream=True,
)

for chunk in stream:
    if chunk.choices:
        print(chunk.choices[0].delta.content or "", end="")

Still haven't found the answer? Try it directly in Playground or view the list of active models.