Implementation:Openai Openai node Legacy Completions Resource

Knowledge Sources	Openai_Openai_node
Domains	SDK, Text Completions
Last Updated	2026-02-15 12:00 GMT

Overview

The legacy Completions class provides access to the OpenAI text completions endpoint (/completions), which generates completions for a given prompt using older models like gpt-3.5-turbo-instruct.

Description

The Completions class at src/resources/completions.ts extends APIResource and wraps the /completions REST endpoint. Unlike the Chat Completions resource, this resource works with raw text prompts rather than structured message arrays. It is accessed via client.completions and provides a single create method with overloads for streaming and non-streaming modes.

The create method accepts a CompletionCreateParams body containing the model and prompt (which can be a string, array of strings, array of tokens, or array of token arrays). The response is a Completion object containing an array of CompletionChoice items, each with text, finish_reason, index, and optional logprobs. The CompletionUsage type tracks token counts including completion_tokens, prompt_tokens, total_tokens, and detailed breakdowns for cached and reasoning tokens.

This resource supports legacy models such as gpt-3.5-turbo-instruct, davinci-002, and babbage-002. Parameters include best_of, echo, frequency_penalty, logit_bias, logprobs, max_tokens, suffix, temperature, top_p, and stop.

Usage

Use this resource for legacy text completion workflows that require raw prompt-in/text-out behavior. For new projects, the Chat Completions API (client.chat.completions) or the Responses API is recommended instead.

Code Reference

Source Location

• Repository: openai-node
• File: src/resources/completions.ts

Signature

export class Completions extends APIResource {
  create(body: CompletionCreateParamsNonStreaming, options?: RequestOptions): APIPromise<Completion>;
  create(body: CompletionCreateParamsStreaming, options?: RequestOptions): APIPromise<Stream<Completion>>;
  create(
    body: CompletionCreateParamsBase,
    options?: RequestOptions,
  ): APIPromise<Stream<Completion> | Completion>;
}

Import

import OpenAI from 'openai';
// Access via client.completions

I/O Contract

Inputs

Name	Type	Required	Description
model	'gpt-3.5-turbo-instruct' | 'davinci-002' | 'babbage-002'	Yes	The model ID to use for completion
prompt	Array<string> | Array<number> | Array<Array<number>> | null	Yes	The prompt(s) to generate completions for
stream	boolean	No	Whether to stream back partial progress
max_tokens	number	No	Maximum number of tokens to generate
temperature	number	No	Sampling temperature between 0 and 2
top_p	number	No	Nucleus sampling parameter
best_of	number	No	Generates best_of completions server-side and returns the best
echo	boolean	No	Echo back the prompt in addition to the completion
suffix	string	No	Suffix that comes after a completion of inserted text
stop	Array<string>	No	Up to 4 sequences where the API will stop generating
logprobs	number	No	Include log probabilities on the most likely output tokens (max 5)

Outputs

Name	Type	Description
Completion	Completion	Object with id, choices (array of CompletionChoice with text, finish_reason), created, model, object ('text_completion'), and optional usage
CompletionChoice	CompletionChoice	'length' | 'content_filter'), index, and logprobs
CompletionUsage	CompletionUsage	Token usage statistics: completion_tokens, prompt_tokens, total_tokens, and detail breakdowns

Usage Examples

import OpenAI from 'openai';

const client = new OpenAI();

// Basic text completion
const completion = await client.completions.create({
  model: 'gpt-3.5-turbo-instruct',
  prompt: 'This is a test.',
  max_tokens: 50,
});
console.log(completion.choices[0].text);

// Streaming text completion
const stream = await client.completions.create({
  model: 'gpt-3.5-turbo-instruct',
  prompt: 'Once upon a time',
  max_tokens: 100,
  stream: true,
});
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0].text);
}

Related Pages

• Environment:Openai_Openai_node_Node_20_Runtime