> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mainly.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Thinking with Nodes
> MainlyAI is a node-based platform for building AI workflows. What does this mean?
export const Node = () => {
const [hovering, setHovering] = useState("");
return <>
It is defined using regular-ish Python code by placing decorators on functions. "Inputs" to the nodes are defined using setHovering("[7]")} onMouseLeave={() => setHovering("")}>@wob.receiver and "outputs" using setHovering("[11]")} onMouseLeave={() => setHovering("")}>@wob.transmitter.
setHovering("")}>
setHovering("[11,12,13]")}>
;
};
## This is a node
*Also known as a Workflow Object, or WOB for short.*
## Nodes are connected by Edges
A transmitter can be connected to multiple receivers, but a receiver can only be connected to one transmitter.
## Edges move data between nodes
Edges move data from transmitters to receivers. Nodes are ran sequentially depth-first.
## Edges care about types
Transmitters and receivers can have different types. Edges can only connect transmitters and receivers of the same type. Each type has a different color.
* **value** - individual values, such as strings, numbers, and vectors.
* **data** - collections of values, such as tables, datasets, and tensors.
* **state** - JSON objects and classes, such as LLM messages and HTTP requests.
* **model** - functions/functors that can be called by nodes later
## Receivers can have controls
Controls are configurable input and display elements for nodes. They can be added onto receivers using the `control` argument. If an edge is connected to a receiver with a control, the control will disappear and the value will be taken from the edge instead.
[Read more about controls](/core_concepts/controls).
```python theme={null}
from mirmod.controls import Textbox
@wob.receiver("value", "input", control=Textbox())
def receive_value(self, value):
self.value = value
```
## Fields let nodes run in loops
They are defined by a pair of receivers and transmitters that are both marked as `is_field=True`. Visually, you can tell that a transmitter or receiver is a field transmitter/receiver by their pointier shape. Nodes within a field are continously run until the field transmitter runs out of values to transmit.
[Read more about fields](/core_concepts/fields).