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

# Using the Anthropic Client

> Run models locally with an Anthropic-style client.

The SDK exposes an Anthropic-style client at `muna.beta.anthropic`, with the same interface as the
official Anthropic client.

<Note>
  The Anthropic client is currently only available in our Python SDK.
</Note>

## Creating Messages

Muna supports running large language models via our client's `anthropic.messages.create` API:

```py Python icon="python" theme={null}
from muna import Muna

# 💥 Create a Muna client
anthropic = Muna().beta.anthropic

# 🔥 Create a message
message = anthropic.messages.create(
  model="@google/gemma-3-270m",
  max_tokens=1024,
  messages=[{ "role": "user", "content": "What is life?" }]
)

# 🚀 Print the result
print(message.content[0].text)
```

## Streaming Messages

Our Anthropic client also supports streaming messages:

```py Python icon="python" theme={null}
from muna import Muna

# 💥 Create a Muna client
anthropic = Muna().beta.anthropic

# 🔥 Stream a message
with anthropic.messages.stream(
  model="@google/gemma-3-270m",
  max_tokens=1024,
  messages=[{ "role": "user", "content": "What is life?" }]
) as stream:
  # 🚀 Use text deltas
  for text in stream.text_stream:
    ...
```

<Tip>
  Call `stream.get_final_message()` to get the complete message once the stream is done.
</Tip>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.