---
title: "模擬 Completion 回應 - 節省測試成本"
url: "/docs/tutorials/mock_completion"
canonical_url: "https://docs.litellm.ai/docs/tutorials/mock_completion"
type: "docs"
last_updated: "2026-10-08"
related:
  - "/docs/tutorials/text_completion"
  - "/docs/tutorials/python_sdk"
---
# 模擬 Completion 回應 - 節省測試成本

> Index of all LiteLLM docs: https://docs.litellm.ai/llms.txt


想要在不呼叫 LLM API 的情況下測試發送 LLM Completion 請求嗎？  
將 `mock_response` 傳給 `litellm.completion`，litellm 就會直接回傳回應，而不需要呼叫 LLM API 並花費 $$ 

## 使用 `completion()` 搭配 `mock_response` {#using-completion-with-mock_response}

```python
from litellm import completion 

model = "gpt-5.6-luna"
messages = [{"role":"user", "content":"Why is LiteLLM amazing?"}]

completion(model=model, messages=messages, mock_response="It's simple to use and easy to get started")
```

## 使用 `completion` 建立 pytest 函式 {#building-a-pytest-function-using-completion}

```python
from litellm import completion
import pytest

def test_completion_openai():
    try:
        response = completion(
            model="gpt-5.6-luna",
            messages=[{"role":"user", "content":"Why is LiteLLM amazing?"}],
            mock_response="LiteLLM is awesome"
        )
        # Add any assertions here to check the response
        print(response)
        print(response['choices'][0]['finish_reason'])
    except Exception as e:
        pytest.fail(f"Error occurred: {e}")
```

## Related pages

- [使用 Text Completion 格式 - 搭配 Completion()](https://docs.litellm.ai/docs/tutorials/text_completion.md)
- [Python SDK](https://docs.litellm.ai/docs/tutorials/python_sdk.md)