multimodal

How is non-textual data passed to a model that has multimodal capabilities?

By providing content blocks.

What three formats do LangChain chat models with multimodal capabilities support for input data?

1) The cross-provider standard format. 2) The OpenAI chat completions format. 3) Any format native to that specific provider (e.g. Anthropic models accept Anthropic's native format).

What happens to the resulting `AIMessage` when a model is invoked to return multimodal data?

It will have content blocks with multimodal types.

What varies from model to model regarding reasoning effort?

The supported effort levels and the provider's documented default.

Where can you check which reasoning effort levels a model supports and its default?

In the model's profile.

In what scenarios is running models locally useful?

When data privacy is critical; when you want to invoke a custom model; or when you want to avoid the costs incurred when using a cloud-based model.

What is described as one of the easiest ways to run chat and embedding models locally?

Ollama.

python response = model.invoke('Create a picture of a cat') print(response.________) # [ # {'type': 'text'; 'text': 'Here's a picture of a cat'}; # {'type': 'image'; 'base64': '...'; 'mime_type': 'image/jpeg'}; # ]

```python response = model.invoke('Create a picture of a cat') print(response.content_blocks) # [ # {'type': 'text'; 'text': 'Here's a picture of a cat'}; # {'type': 'image'; 'base64': '...'; 'mime_type': 'image/jpeg'}; # ] ```

python model.profile['________'] # e.g. ['low'; 'medium'; 'high'] model.profile['reasoning_effort_default'] # e.g. 'high'

```python model.profile['reasoning_effort_levels'] # e.g. ['low'; 'medium'; 'high'] model.profile['reasoning_effort_default'] # e.g. 'high' ```

python model.profile['reasoning_effort_levels'] # e.g. ['low'; 'medium'; 'high'] model.profile['________'] # e.g. 'high'

```python model.profile['reasoning_effort_levels'] # e.g. ['low'; 'medium'; 'high'] model.profile['reasoning_effort_default'] # e.g. 'high' ```

python from langchain_anthropic import ChatAnthropic model = ________(model='claude-sonnet-4-6') response = model.invoke( 'Why do parrots have colorful feathers?'; ________='high'; )

```python from langchain_anthropic import ChatAnthropic model = ChatAnthropic(model='claude-sonnet-4-6') response = model.invoke( 'Why do parrots have colorful feathers?'; reasoning_effort='high'; ) ```

python response = model.invoke('Create a picture of a cat') print(response.content_blocks) # [ # {'type': 'text'; 'text': 'Here's a picture of a cat'}; # {'type': '________'; 'base64': '...'; 'mime_type': 'image/jpeg'}; # ]

```python response = model.invoke('Create a picture of a cat') print(response.content_blocks) # [ # {'type': 'text'; 'text': 'Here's a picture of a cat'}; # {'type': 'image'; 'base64': '...'; 'mime_type': 'image/jpeg'}; # ] ```