> ## Documentation Index
> Fetch the complete documentation index at: https://docs.morph-data.io/llms.txt
> Use this file to discover all available pages before exploring further.

# create_chunk

This function creates a stream response for LLM Apps.
By using the `MorphChatStreamChunk` returned by this function and passing it to `<LLM/>`, you can build a chat application utilizing LLM.

```python morph_lib.stream theme={"dark"}
# 📦 Package: morph_lib.stream
# 🛠️ Function: create_chunk

def create_chunk(
    text: Optional[str] = None, content: Optional[str] = None
) -> MorphChatStreamChunk:
```

### Parameters

<ParamField body="text" type="str">
  Text to be displayed in the chat stream.
</ParamField>

<ParamField body="content" type="str">
  Content to be displayed in the side-by-side layout of `<LLM/>`.

  ex.) html, markdown, etc.
</ParamField>

### Example

```python theme={"dark"}
import os
from morph_lib.stream import create_chunk
from openai import OpenAI
import morph
from morph import MorphGlobalContext

@morph.func(name="test")
@morph.variables("prompt")
def main(context: MorphGlobalContext):
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    propmt = context.vars["prompt"]
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": propmt}],
        stream=True,
    )

    for c in response:
        text = c.model_dump()["choices"][0]["delta"].get("content", "")
        yield create_chunk(text=text)
```