Interface FakeStreamingChatModelFields

Interface for the Constructor-field specific to the Fake Streaming Chat model (all optional because we fill in defaults).

interface FakeStreamingChatModelFields {
    cache?: boolean | BaseCache<Generation[]>;
    callbackManager?: CallbackManager;
    callbacks?: Callbacks;
    chunks?: AIMessageChunk[];
    disableStreaming?: boolean;
    maxConcurrency?: number;
    maxRetries?: number;
    metadata?: Record<string, unknown>;
    onFailedAttempt?: FailedAttemptHandler;
    responses?: BaseMessage[];
    sleep?: number;
    tags?: string[];
    thrownErrorString?: string;
    toolStyle?:
        | "openai"
        | "anthropic"
        | "bedrock"
        | "google";
    verbose?: boolean;
}

Hierarchy (view full)

Properties

cache? callbackManager? callbacks? chunks? disableStreaming? maxConcurrency? maxRetries? metadata? onFailedAttempt? responses? sleep? tags? thrownErrorString? toolStyle? verbose?

Properties

Optionalcache

cache?: boolean | BaseCache<Generation[]>

OptionalcallbackManager

callbackManager?: CallbackManager

Deprecated

Use callbacks instead

Optionalcallbacks

callbacks?: Callbacks

Optionalchunks

chunks?: AIMessageChunk[]

Exact chunks to emit (can include tool-call deltas)

OptionaldisableStreaming

disableStreaming?: boolean

Whether to disable streaming.

If streaming is bypassed, then stream() will defer to invoke().

  • If true, will always bypass streaming case.
  • If false (default), will always use streaming case if available.

OptionalmaxConcurrency

maxConcurrency?: number

The maximum number of concurrent calls that can be made. Defaults to Infinity, which means no limit.

OptionalmaxRetries

maxRetries?: number

The maximum number of retries that can be made for a single call, with an exponential backoff between each attempt. Defaults to 6.

Optionalmetadata

metadata?: Record<string, unknown>

OptionalonFailedAttempt

onFailedAttempt?: FailedAttemptHandler

Custom handler to handle failed attempts. Takes the originally thrown error object as input, and should itself throw an error if the input error is not retryable.

Optionalresponses

responses?: BaseMessage[]

Full AI messages to fall back to when no chunks supplied

Optionalsleep

sleep?: number

Milliseconds to pause between fallback char-by-char chunks

Optionaltags

tags?: string[]

OptionalthrownErrorString

thrownErrorString?: string

Throw this error instead of streaming (useful in tests)

OptionaltoolStyle

toolStyle?:
    | "openai"
    | "anthropic"
    | "bedrock"
    | "google"

How tool specs are formatted in bindTools

Optionalverbose

verbose?: boolean

Settings

Member Visibility

On This Page

Properties