org.llm4s.llmconnect.utils

Members list

Type members

Classlikes

object ChunkingUtils

Utilities for splitting text, audio, and video data into fixed-size overlapping chunks.

Utilities for splitting text, audio, and video data into fixed-size overlapping chunks.

Each chunking method produces a sequence of windows with configurable size and overlap, suitable for embedding pipelines and multimodal processing.

Attributes

Supertypes
class Object
trait Matchable
class Any
Self type
final case class ClientHealth(status: ConnectionStatus, lastHealthCheck: Instant, responseTimeMs: Option[Long], errorCount: Long, capabilities: Option[ProviderCapabilities])

Health check snapshot for an LLM client connection.

Health check snapshot for an LLM client connection.

Value parameters

capabilities

provider capabilities discovered during the health check

errorCount

cumulative number of errors observed on this connection

lastHealthCheck

timestamp of the most recent health check

responseTimeMs

round-trip time of the last successful health check in milliseconds

status

current connection status

Attributes

Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all
sealed trait ConnectionStatus

Connection status for LLM clients.

Connection status for LLM clients.

Represents the current state of a provider connection: connected, disconnected, connecting, or in an error state with an optional underlying cause.

Attributes

Companion
object
Supertypes
class Object
trait Matchable
class Any
Known subtypes
object Connected
object Connecting
object Disconnected
class Error

Attributes

Companion
trait
Supertypes
trait Sum
trait Mirror
class Object
trait Matchable
class Any
Self type
object ModelSelector

Selects an embedding model configuration based on input modality.

Selects an embedding model configuration based on input modality.

Text modality selection is intentionally disallowed here; callers should load text embedding models through their typed config layer. Image, audio, and video modalities are resolved from the supplied LocalEmbeddingModels config.

Attributes

Supertypes
class Object
trait Matchable
class Any
Self type
final case class ProviderCapabilities(supportsStreaming: Boolean, supportsToolCalls: Boolean, supportsFunctionCalling: Boolean, supportsVision: Boolean, maxTokens: Option[Int], supportedModels: List[String], metadata: Map[String, String])

Describes the capabilities supported by an LLM provider.

Describes the capabilities supported by an LLM provider.

Value parameters

maxTokens

maximum token limit for completions, if known

metadata

additional provider-specific key-value metadata

supportedModels

list of model identifiers the provider offers

supportsFunctionCalling

whether the provider supports legacy function calling

supportsStreaming

whether the provider supports streaming responses

supportsToolCalls

whether the provider supports tool/function calling

supportsVision

whether the provider supports image/vision inputs

Attributes

Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all

Vector similarity utilities for embedding comparison.

Vector similarity utilities for embedding comparison.

Attributes

Supertypes
class Object
trait Matchable
class Any
Self type