Completions

The legacy text completion endpoint generates text from an input prompt.

Recommendation

For most conversational use cases, use the Chat API. It supports multi-turn conversations and provides more complete functionality.

Create a completion

POST https://api.nexusmodels.cn/v1/completions

Request parameters

Parameter	Type	Required	Description
model	string	Yes	Model ID
prompt	string/array	Yes	Input prompt
suffix	string	No	Text appended after the generated content
max_tokens	integer	No	Maximum number of generated tokens
temperature	number	No	Sampling temperature, usually between 0 and 2
top_p	number	No	Nucleus sampling value, usually between 0 and 1
n	integer	No	Number of completion candidates to return
stream	boolean	No	Enables streaming output
logprobs	integer	No	Number of token log probabilities to return
echo	boolean	No	Includes the original prompt in the response
stop	string/array	No	Sequence that stops generation
presence_penalty	number	No	Presence penalty
frequency_penalty	number	No	Frequency penalty
best_of	integer	No	Generates multiple results on the server and returns the best one

Parameter compatibility

Supported parameters may vary by model. Unsupported parameters may be ignored or cause the request to fail.

cURL example

curl -sS -X POST \
  'https://api.nexusmodels.cn/v1/completions' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_NEXUSMODELS_API_KEY' \
  -d '{
    "model": "YOUR_COMPLETION_MODEL",
    "prompt": "Write a short greeting:",
    "max_tokens": 50,
    "temperature": 0.7
  }'

Response

A successful request returns:

200 OK

{
  "id": "cmpl-abc123",
  "object": "text_completion",
  "created": 1788163200,
  "model": "YOUR_COMPLETION_MODEL",
  "choices": [
    {
      "text": "Hello! I hope you have a wonderful and productive day.",
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 18,
    "total_tokens": 33
  }
}

Response fields

Field	Description
id	Unique identifier for the completion request
object	Response object type
created	Response creation timestamp
model	Model used for the request
choices	Completion candidates generated by the model
choices[].text	Generated text
choices[].index	Candidate index
choices[].logprobs	Token log probabilities
choices[].finish_reason	Reason generation stopped
usage	Token usage information for the request

Code completion example

This example requires a model that supports code completion.

curl -sS -X POST \
  'https://api.nexusmodels.cn/v1/completions' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_NEXUSMODELS_API_KEY' \
  -d '{
    "model": "YOUR_COMPLETION_MODEL",
    "prompt": "// JavaScript: calculate the sum of all even numbers in an array\nfunction sumEvenNumbers(arr) {\n",
    "max_tokens": 100,
    "temperature": 0.2
  }'

Python example

Install the dependency:

pip install requests

Call the API:

import os
import requests