X (Twitter)

Docs

Completions

Use the completions endpoint for text completion tasks.

The completions endpoint generates text completions given a prompt. For most use cases, we recommend using chat completions instead.

Endpoint

POST https://api.elyxir.ai/v1/completions

When to Use

Use this endpoint for:

Legacy applications using the completions API
Simple text completion without conversation context
Compatibility with older OpenAI SDK versions

For new applications, prefer /v1/chat/completions.

Request

Headers

Header	Required	Description
Authorization	Yes	Bearer token: `Bearer elyxir_xxx`
Content-Type	Yes	Must be `application/json`

Body Parameters

Parameter	Type	Required	Description
model	string	Yes	Model ID
prompt	string/array	Yes	Text to complete
max_tokens	integer	No	Maximum tokens to generate
temperature	number	No	Randomness (0-2, default: 1)
top_p	number	No	Nucleus sampling (0-1)
stop	string/array	No	Stop sequences
echo	boolean	No	Echo prompt in response

Example Request

curl -X POST https://api.elyxir.ai/v1/completions \
  -H "Authorization: Bearer elyxir_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-3.5-turbo-instruct",
    "prompt": "Write a haiku about programming:",
    "max_tokens": 50,
    "temperature": 0.7
  }'

Response

Success (200 OK)

{
  "id": "cmpl-abc123",
  "object": "text_completion",
  "created": 1234567890,
  "model": "gpt-3.5-turbo-instruct",
  "choices": [
    {
      "text": "\nCode flows like water\nBugs emerge from the depths\nDebug, iterate",
      "index": 0,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 6,
    "completion_tokens": 20,
    "total_tokens": 26
  }
}

Multiple Completions

Request multiple completions with the n parameter:

{
  "model": "gpt-3.5-turbo-instruct",
  "prompt": "Generate a product name for a coffee shop:",
  "n": 3,
  "max_tokens": 20
}

Response includes multiple choices:

{
  "choices": [
    {"text": "Bean & Brew", "index": 0},
    {"text": "Morning Roast", "index": 1},
    {"text": "The Daily Grind", "index": 2}
  ]
}

Available Models

Not all models support the completions endpoint. Compatible models:

Model	Provider
gpt-3.5-turbo-instruct	OpenAI
davinci-002	OpenAI
babbage-002	OpenAI

Most modern models only support chat completions. Check model documentation for compatibility.

Migrating to Chat Completions

Convert your prompt to a messages array:

Before (Completions):

{
  "model": "gpt-3.5-turbo-instruct",
  "prompt": "Write a haiku about programming:"
}

After (Chat Completions):

{
  "model": "gpt-4o-mini",
  "messages": [
    {"role": "user", "content": "Write a haiku about programming:"}
  ]
}

Chat Completions - Recommended endpoint
API Overview - Getting started
Error Handling - Handle errors

Chat Completions Embeddings