Text & embeddings

Completions (legacy)

Generate text from a single prompt with the legacy OpenAI Completions API.

POST /v1/completions

The legacy OpenAI Completions API: the model continues a plain-text prompt. It is kept for older tools and for models that only support completion-style input (e.g. gpt-3.5-turbo-instruct, code completion with suffix).

For new projects use Chat Completions or Responses — most current models do not support this endpoint.

Authentication

Authorization: Bearer sk-xxxx

Request body

Parameter	Type	Required	Description
`model`	string	Yes	Model ID, e.g. `gpt-3.5-turbo-instruct`.
`prompt`	string \| array	Yes	Prompt text, or an array of prompts.
`suffix`	string	No	Text that follows the completion (fill-in-the-middle).
`max_tokens`	integer	No	Maximum number of tokens to generate. Default `16`.
`temperature`	number	No	Sampling temperature, `0`–`2`.
`top_p`	number	No	Nucleus sampling, `0`–`1`.
`n`	integer	No	Number of completions per prompt.
`stop`	string \| array	No	Up to 4 stop sequences.
`presence_penalty`	number	No	`-2.0`–`2.0`.
`frequency_penalty`	number	No	`-2.0`–`2.0`.
`logprobs`	integer	No	Return log probabilities of the top N tokens.
`seed`	integer	No	Best-effort deterministic sampling.
`stream`	boolean	No	Stream as SSE, ending with `data: [DONE]`. See Streaming.
`stream_options`	object	No	`{"include_usage": true}` adds a final usage chunk.
`user`	string	No	Stable end-user identifier.

Request example

cURL

curl https://rutoken.ai/v1/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $API_KEY" \
  -d '{
    "model": "gpt-3.5-turbo-instruct",
    "prompt": "Write a tagline for a coffee shop:",
    "max_tokens": 30,
    "temperature": 0.8
  }'

Python

from openai import OpenAI

client = OpenAI(base_url="https://rutoken.ai/v1", api_key="sk-xxxx")
resp = client.completions.create(
    model="gpt-3.5-turbo-instruct",
    prompt="Write a tagline for a coffee shop:",
    max_tokens=30,
    temperature=0.8,
)
print(resp.choices[0].text)

Node.js

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://rutoken.ai/v1", apiKey: "sk-xxxx" });
const resp = await client.completions.create({
  model: "gpt-3.5-turbo-instruct",
  prompt: "Write a tagline for a coffee shop:",
  max_tokens: 30,
  temperature: 0.8,
});
console.log(resp.choices[0].text);

Response

json

{
  "id": "cmpl-9xYz",
  "object": "text_completion",
  "created": 1760000000,
  "model": "gpt-3.5-turbo-instruct",
  "choices": [
    {
      "text": " Brewed with love, served with a smile.",
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 10,
    "total_tokens": 19
  }
}

Field	Type	Description
`choices[].text`	string	Generated text.
`choices[].finish_reason`	string	`stop` or `length`.
`usage`	object	Token counts used for billing.

← Previous

Gemini generateContent

Embeddings