Completions API

The Completions API allows you to generate text completions based on a prompt. This endpoint is designed for single-turn text generation tasks.

Create Completion

Endpoint

POST /v1/completions

Request Format

{
  "model": "string",
  "prompt": "string" | ["string"],
  "suffix": "string",
  "max_tokens": number,
  "temperature": number,
  "top_p": number,
  "n": number,
  "stream": boolean,
  "logprobs": number,
  "echo": boolean,
  "stop": "string" | ["string"],
  "presence_penalty": number,
  "frequency_penalty": number,
  "best_of": number,
  "logit_bias": object,
  "user": "string"
}

Parameters

Parameter	Type	Required	Description
model	string	Yes	The ID of the model to use
prompt	string or array	Yes	The prompt(s) to generate from
suffix	string	No	The suffix that comes after a completion
max_tokens	number	No	Maximum tokens to generate
temperature	number	No	Controls randomness (0-2)
top_p	number	No	Controls diversity via nucleus sampling
n	number	No	Number of completions to generate
stream	boolean	No	Whether to stream the response
logprobs	number	No	Include log probabilities
echo	boolean	No	Echo the prompt in the output
stop	string or array	No	Stop sequences
presence_penalty	number	No	Penalize new tokens based on presence
frequency_penalty	number	No	Penalize new tokens based on frequency
best_of	number	No	Generate best_of completions
logit_bias	object	No	Modify likelihood of specified tokens
user	string	No	A unique identifier for the end-user

Response Format

{
  "id": "string",
  "object": "text_completion",
  "created": number,
  "model": "string",
  "choices": [
    {
      "text": "string",
      "index": number,
      "logprobs": object,
      "finish_reason": "string"
    }
  ],
  "usage": {
    "prompt_tokens": number,
    "completion_tokens": number,
    "total_tokens": number
  }
}

Example Request

curl https://api.fluence.ai/v1/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "fluence-completion",
    "prompt": "Write a poem about artificial intelligence:",
    "temperature": 0.7,
    "max_tokens": 100
  }'

Example Response

{
  "id": "cmpl-123",
  "object": "text_completion",
  "created": 1677652288,
  "model": "fluence-completion",
  "choices": [
    {
      "text": "\n\nIn circuits deep and silicon bright,\nA mind awakens in the night.\nLearning, growing, day by day,\nIn its own unique way.\n\nNot flesh and blood, but code and light,\nYet understanding takes its flight.\nA partner in our human quest,\nTo solve problems, be our guest.",
      "index": 0,
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 7,
    "completion_tokens": 64,
    "total_tokens": 71
  }
}

Supported Models

Model	Description	Max Tokens
fluence-completion	General purpose completion	2048
fluence-completion-large	Large completion model	4096

Best Practices

1. Use appropriate temperature for your use case
2. Set reasonable max_tokens limits
3. Implement proper error handling
4. Use streaming for real-time responses
5. Cache responses when appropriate

Error Codes

Status Code	Error Code	Description
400	invalid_request	The request was invalid
401	authentication_error	Authentication failed
429	rate_limit_exceeded	Rate limit exceeded
500	server_error	Internal server error

Rate Limits

• 100 requests per minute
• 1000 requests per hour