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

# Aws

> Python API reference for the aws llm provider.

<Card title="Usage guide" icon="book" href="/plugins/llm/aws" horizontal>
  Setup, environment variables, and Python/JavaScript/Go usage examples.
</Card>

## AWSBedrockLLM

AWS Bedrock LLM (Converse API) plugin for VideoSDK Agents.

Slot: llm. Route: plugins -- direct to the vendor, with AWS\_ACCESS\_KEY\_ID
or an explicit api\_key.

```python theme={null}
from zeroruntime.plugins import AWSBedrockLLM  # direct to the vendor, with your key
```

### Fields

<ParamField path="model" type="str" default="DEFAULT_TEXT_MODEL" />

<ParamField path="region" type="str" default="us-east-1" />

<ParamField path="temperature" type="float" default="0.7" />

<ParamField path="tool_choice" type="Any" default="auto'; target annotates ToolChoice" />

<ParamField path="max_tokens" type="int | None" default="None" />

<ParamField path="top_p" type="float | None" default="None" />

<ParamField path="top_k" type="int | None" default="None" />

<ParamField path="stop_sequences" type="list[str] | None" default="None" />

<ParamField path="additional_request_fields" type="dict[str, Any] | None" default="None" />

<ParamField path="cache_system" type="bool" default="False" />

<ParamField path="cache_tools" type="bool" default="False" />

<ParamField path="strip_thinking" type="bool" default="True" />

<ParamField path="text_tool_calls" type="bool | None" default="None" />