FakeEmbeddingModel class final

An EmbeddingModel returning deterministic vectors.

Vectors are derived from the text's characters, so identical text embeds identically and different text embeds differently — enough to test a retrieval pipeline end to end without a network or a real model.

Implemented types
Available extensions

Constructors

FakeEmbeddingModel({int dimensions = 8, int maxBatchSize = 100})
Creates a fake embedder.

Properties

dimensions → int
Number of components each vector has.
final
embedded → List<String>
Every input this model was asked to embed.
final
hashCode → int
The hash code for this object.
no setterinherited
info → ModelInfo
What this model is.
final
maxBatchSize → int
Maximum inputs accepted in one call.
final
runtimeType → Type
A representation of the runtime type of the object.
no setterinherited

Methods

dispose() → Future<void>
Releases every resource held by this object.
embed(List<String> inputs, {required EmbeddingPurpose purpose, AgenticContext? context}) → Future<List<Embedding>>
Embeds inputs, returning one vector per input, in order.
override
embedAll(List<String> inputs, {required EmbeddingPurpose purpose, AgenticContext? context, void onProgress(int done, int total)?}) → Future<List<Embedding>>

Available on EmbeddingModel, provided by the EmbeddingModelOperations extension

Embeds any number of inputs, splitting into provider-sized batches.
embedDocument(String text, {AgenticContext? context}) → Future<Embedding>

Available on EmbeddingModel, provided by the EmbeddingModelOperations extension

Embeds a single document.
embedQuery(String text, {AgenticContext? context}) → Future<Embedding>

Available on EmbeddingModel, provided by the EmbeddingModelOperations extension

Embeds a single search query.
noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited