IMPORTANT: In July 2025, we stopped releasing new versions of KTX modules, and we removed the KTX libraries from the Firebase Android BoM (v34.0.0). If you use KTX APIs from the previously released KTX modules, we recommend that you migrate your app to use KTX APIs from the main modules instead. For details, see the FAQ about this initiative.

InferenceMode

@PublicPreviewAPI
class InferenceMode

Specifies how the SDK should choose between on-device and in-cloud inference.

Summary

Public companion properties
`InferenceMode`	`ONLY_IN_CLOUD` Only use in-cloud inference.
`InferenceMode`	`ONLY_ON_DEVICE` Only use on-device inference.
`InferenceMode`	`PREFER_IN_CLOUD` Prefer in-cloud inference, but fallback to on-device if cloud is unavailable.
`InferenceMode`	`PREFER_ON_DEVICE` Prefer on-device inference, but fallback to in-cloud if unavailable.

Public companion properties

ONLY_IN_CLOUD

val ONLY_IN_CLOUD: InferenceMode

Only use in-cloud inference.

ONLY_ON_DEVICE

val ONLY_ON_DEVICE: InferenceMode

Only use on-device inference.

PREFER_IN_CLOUD

val PREFER_IN_CLOUD: InferenceMode

Prefer in-cloud inference, but fallback to on-device if cloud is unavailable.

In this mode, the SDK will use in-cloud inference only unless the device is offline, at which point the SDK will fall back to on-device inference.

PREFER_ON_DEVICE

val PREFER_ON_DEVICE: InferenceMode

Prefer on-device inference, but fallback to in-cloud if unavailable.

In this mode, the SDK will attempt to use on-device inference first and, if the on-device model is unable to generate an answer, it will retry using the cloud model.