Chat Completions

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "plm_xxxxxxxxxxxxxxxx",
  baseURL: "https://apirouter.pleum.ai/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-5.5",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

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

from openai import OpenAI

client = OpenAI(
    api_key="plm_xxxxxxxxxxxxxxxx",
    base_url="https://apirouter.pleum.ai/v1",
)

response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[{"role": "user", "content": "Why is the sky blue?"}],
)

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

curl https://apirouter.pleum.ai/v1/chat/completions \
  -H "Authorization: Bearer plm_xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "messages": [{"role": "user", "content": "Why is the sky blue?"}]
  }'

Request body#

The standard sampling / tool parameters (top_p through reasoning_effort) are passed through to the routed provider as-is, without validation — a parameter the model does not support may be ignored by the provider or cause it to return an error. n (multiple choices) is not supported — a single response is always returned.

{
  "model": "gpt-5.5",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Hello"}
  ],
  "temperature": 0.7,
  "max_tokens": 4096,
  "stream": false
}

Response#

Returns OpenAI-style choices / usage. Cost is settled after the reply is sent, so cost (KRW cost, FX rate, markup) is included in the body only once settlement has finished. If it is missing, call GET /v1/generation?id=… with the X-Request-Id response header value, or check dashboard usage. See also Where to find cost.

{
  "id": "chatcmpl-9f3c2b7a5e1d4c8a9b0e6d7f1a2b3c4d",
  "object": "chat.completion",
  "created": 1767225600,
  "model": "gpt-5.5",
  "choices": [
    {
      "index": 0,
      "message": {"role": "assistant", "content": "Hello! How can I help you?"},
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 24,
    "completion_tokens": 12,
    "total_tokens": 36
  },
  "cost": {"usd": 0.00048, "krw": 0.7, "fx_rate": 1342.48, "markup_rate": 0.0}
}

Streaming#

With stream: true, text chunks are sent as text/event-stream, and the final event carries usage. Look up cost with X-Request-Id.

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"Hello! "}}]}

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"How can I help you?"}}]}

data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: {"choices":[],"usage":{"prompt_tokens":12,"completion_tokens":8,"total_tokens":20}}

data: [DONE]

Prompt caching#

PleumRouter supports prompt caching. OpenAI, Google Gemini, and DeepSeek cache automatically with no extra configuration — repeated input prefixes (long system prompts, reference documents, etc.) are cached on the provider side, and cache hits are billed at a discounted rate below the standard input price, lowering your cost.

Anthropic (Claude) uses explicit caching. Add cache_control: {"type": "ephemeral"} to a content part to mark a cache breakpoint; the prefix up to that point is cached. On multi-turn calls, breakpoints are injected automatically even without cache_control; see Auto Prompt Caching for the rules.

{
  "model": "claude-sonnet-4-6",
  "messages": [
    {
      "role": "system",
      "content": [
        {
          "type": "text",
          "text": "<large reusable context: docs, schema, instructions...>",
          "cache_control": {"type": "ephemeral"}
        }
      ]
    },
    {"role": "user", "content": "Answer based on the context above."}
  ]
}

On a cache hit, the response usage includes prompt_tokens_details.cached_tokens (hit tokens). Anthropic also returns cache-write tokens as cache_creation_input_tokens. The discount applies to models that list a cache-read price in cache_read_usd on GET /v1/models.

"usage": {
  "prompt_tokens": 10240,
  "completion_tokens": 120,
  "total_tokens": 10360,
  "prompt_tokens_details": {"cached_tokens": 10000},
  "cache_creation_input_tokens": 0
}

Billing#

A call that fails before the response starts is not billed. A successful call is settled against actual token usage, with a minimum charge of 0.1 credits. If a stream is cut off midway, the tokens already sent are billed.