---
title: "Completion"
slug: "completion"
category: "glossary"
tags: ["glossary", "ai-agents", "completion", "llm", "generation", "token"]
status: "stable"
last_updated: 2026-08-14
summary: "Completion is the text a language model generates in response to a prompt, a term from the original text-completion API paradigm."
related:
  [
    "[[glossary/token]]",
    "[[glossary/stop-sequence]]",
    "[[glossary/system-prompt]]",
    "[[glossary/temperature]]",
    "[[glossary/prompt-cache]]",
    "[[prompt-engineering/prompt-design]]",
  ]
---

## Overview

This page is the atomic definition. The prompting deep-dive lives at [[prompt-engineering/prompt-design]]. A completion is the generated output an LLM returns for a given input, produced one token at a time and terminated by an end-of-turn signal, a stop sequence, or a token limit.

## Definition

A completion is the generated output returned by a language model given an input prompt. In the original completion API paradigm (GPT-3, early OpenAI), the model received raw text and generated a continuation. Modern APIs use a chat/messages format where input is structured as a list of turns, but the output is still called a completion.

A completion consists of one or more content blocks. In most APIs:
- `text` blocks contain generated text.
- `tool_use` blocks contain structured function calls (see [[glossary/tool-call]]).

Key completion parameters:
- `max_tokens`: upper bound on output length in [[glossary/token|tokens]].
- `temperature`: controls randomness; 0 is near-deterministic, 1 is sampling from the full distribution.
- `top_p`: nucleus sampling threshold; an alternative to temperature.
- `stop_sequences`: strings that halt generation early (see [[glossary/stop-sequence]]).

The model generates one token at a time, sampling from a probability distribution over the vocabulary at each step. The sequence of tokens produced is the completion. The model cannot revise earlier tokens once generated.

`stop_reason` indicates why generation ended: `"end_turn"` (model decided to stop), `"max_tokens"` (hit the limit), `"stop_sequence"` (a stop sequence matched).

## When it applies

Treat every completion as potentially truncated: check `stop_reason`. If `stop_reason = "max_tokens"`, the output was cut off; increase `max_tokens` or break the task into smaller parts. For deterministic pipelines (classification, structured extraction) set `temperature = 0`. For creative generation set `temperature` between 0.7 and 1.0.

## Example

```python
import anthropic

client = anthropic.Anthropic()
message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Explain backpressure in one paragraph."}]
)

print(f"Stop reason: {message.stop_reason}")
print(message.content[0].text)
```

## Related concepts

- [[glossary/token]] - completions are generated one token at a time.
- [[glossary/stop-sequence]] - stop sequences bound completion length semantically.
- [[glossary/temperature]] - temperature controls the randomness of each token selection.
- [[glossary/prompt-cache]] - prompt caching reduces the cost of repeated completions over the same context.
- [[prompt-engineering/prompt-design]] - the prompting deep-dive on eliciting useful completions.

## Citing this term

> See [[glossary/completion|Completion]] (llmbestpractices.com/glossary/completion).

## Related

- [[glossary/token]]
- [[glossary/stop-sequence]]
- [[glossary/system-prompt]]
- [[glossary/temperature]]
- [[prompt-engineering/prompt-design]]