OpenAIBackend class

OpenAI (or any OpenAI-compatible endpoint) via stock dartantic OpenAIProvider / OpenAIChatModel.

Note: the OpenAI path force-wraps a transport-only RetryHttpClient (429 / 5xx / IO). That retry does NOT re-sample on a schema rejection (a 200 with bad content), so the decision-retry contract is intact; it only removes driver control over transport retries on this backend.

Inheritance

• Object
• ModelBackendSpec
• OpenAIBackend

Constructors

OpenAIBackend({required String apiKey, Uri? baseUrl, Map<String, String>? headers, bool enableThinking = false, double? temperature, OpenAIChatOptions? options})

Creates an OpenAI backend spec.

const

Properties

apiKey → String

OpenAI API key.

final

baseUrl → Uri?

Override endpoint (e.g. an OpenAI-compatible gateway /v1).

final

enableThinking → bool

Whether reasoning is requested (only honoured by reasoning-capable models; the OpenAI Chat Completions path does not surface thinking).

final

hashCode → int

The hash code for this object.

no setter, inherited

headers → Map<String, String>?

Extra headers.

final

options → OpenAIChatOptions?

OpenAI sampling options.

final

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

temperature → double?

Sampling temperature.

final

Methods

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