## VariadicSplit {#openvino_docs_ops_movement_VariadicSplit_1}
**Versioned name**: *VariadicSplit-1*
**Category**: *Data movement*
**Short description**: *VariadicSplit* operation splits an input tensor into chunks along some axis. The chunks may have variadic lengths depending on `split_lengths` input tensor.
**Detailed Description**
*VariadicSplit* operation splits a given input tensor `data` into chunks along a scalar or tensor with shape `[1]` `axis`. It produces multiple output tensors based on additional input tensor `split_lengths`.
The i-th output tensor shape is equal to the input tensor `data` shape, except for dimension along `axis` which is `split_lengths[i]`.
\f[
shape\_output\_tensor = [data.shape[0], data.shape[1], \dotsc , split\_lengths[i], \dotsc , data.shape[D-1]]
\f]
Where D is the rank of input tensor `data`. The sum of elements in `split_lengths` must match `data.shape[axis]`.
**Attributes**: *VariadicSplit* operation has no attributes.
**Inputs**
* **1**: `data`. A tensor of type `T1` and arbitrary shape. **Required.**
* **2**: `axis`. Axis along `data` to split. A scalar or tensor with shape `[1]` of type `T2` with value from range `-rank(data) .. rank(data)-1`. Negative values address dimensions from the end.
**Required.**
* **3**: `split_lengths`. A list containing the dimension values of each output tensor shape along the split `axis`. A 1D tensor of type `T2`. The number of elements in `split_lengths` determines the number of outputs. The sum of elements in `split_lengths` must match `data.shape[axis]`. In addition `split_lengths` can contain a single `-1` element, which means, all remaining items along specified `axis` that are not consumed by other parts. **Required.**
**Outputs**
* **Multiple outputs**: Tensors of type `T1`. The i-th output has the same shape as `data` input tensor except for dimension along `axis` which is `split_lengths[i]` if `split_lengths[i] != -1`. Otherwise, the dimension along `axis` is processed as described in `split_lengths` input description.
**Types**
* *T1*: any arbitrary supported type.
* *T2*: any integer type.
**Examples**
```xml
61210243
```
```xml
61210242
```