---
title: CircuitItemModel (latest version)
description: API reference for ibm_quantum_schemas.executor.version_2_0.CircuitItemModel in the latest version of qiskit-ibm-runtime
source: https://eu-de.quantum.cloud.ibm.com/docs/en/api/qiskit-ibm-runtime/ibm-quantum-schemas-executor-version-2-0-circuit-item-model
---

# CircuitItemModel

*pydantic model* `CircuitItemModel`

[GitHub](https://github.com/Qiskit/ibm-quantum-schemas/tree/0.10.20260723/ibm_quantum_schemas/executor/version_2_0/models.py#L101-L131)

Bases: `BaseModel`

Execution specifications for a single quantum circuit.

The circuit for each item is store separately in `QuantumProgramModel`.

### chunk\_size

*field*

Type: `Annotated`\[`int`, `Field`(`ge``=``1`)] | `Literal`\[`'auto'`]

Default value: `'auto'`

The maximum number circuit arguments to bind to the circuit per shot loop.

When `"auto"`, the number is chosen server-side with heuristics designed to optimize execution speed. A quantum program must have items where either all chunk sizes are integer-valued, or all chunk sizes are `"auto"`. Integer values are only allowed inside of session exection mode.

### circuit\_arguments

*field*

Type: [`F64CompressedTensorModel`](/docs/api/qiskit-ibm-runtime/ibm-quantum-schemas-common-f-64-compressed-tensor-model "ibm_quantum_schemas.common.F64CompressedTensorModel")

Arguments to the parameters of the circuit.

The last axis is over `circuit.parameters`. Execution broadcasts over the preceding axes; expect one result per element of the leading shape.

### item\_type

*field*

Type: `Literal`\[`'circuit'`]

Default value: `'circuit'`

The type of quantum program item.

### shape

*field*

Type: `list`\[`int`]

The shape of this item.

This shape must extend (via broadcasting) the implicit shape of the :attr:\~circuit\_arguments\`. The non-trivial axes it introduces represent replications.