llm4s-docs
API
Generated with
llm4s-docs/org.llm4s/org.llm4s.llmconnect/org.llm4s.llmconnect.config/CohereConfig

CohereConfig

org.llm4s.llmconnect.config.CohereConfig
See theCohereConfig companion class
@Experimental
object CohereConfig

Attributes

Companion
class
Graph
Supertypes
trait Product
trait Mirror
class Object
trait Matchable
class Any
Self type

Members list

Type members

Inherited and Abstract types

type MirroredElemLabels <: Tuple

The names of the product elements

The names of the product elements

Attributes

Inherited from:
Mirror
type MirroredLabel <: String

The name of the type

The name of the type

Attributes

Inherited from:
Mirror

Value members

Concrete methods

def apply(apiKey: String, model: String, baseUrl: String, contextWindow: Int, reserveCompletion: Int, timeouts: ProviderTimeouts): CohereConfig

Builds a config without validating it; fromValues validates and resolves the context window from the model. Every field after model has a default.

Builds a config without validating it; fromValues validates and resolves the context window from the model. Every field after model has a default.

Attributes

def apply(apiKey: String, model: String): CohereConfig

The API key and model, every other field at its default: the entry point for Java and Kotlin, which do not see Scala default arguments. Set the rest with the with* setters.

The API key and model, every other field at its default: the entry point for Java and Kotlin, which do not see Scala default arguments. Set the rest with the with* setters.

Attributes

def compatibilityBaseUrl(baseUrl: String): String

The Compatibility API base for a configured base URL.

The Compatibility API base for a configured base URL.

A URL already ending in /compatibility/v1 is used as it is. Anything else is taken to be a Cohere API root - which is what baseUrl meant before #1132, when the client appended /v2/chat to it - and gets /compatibility/v1 appended, after dropping a trailing native version segment (/v1 or /v2). So https://api.cohere.com, https://api.cohere.com/v2 and a proxy root that forwards to Cohere all keep working. Idempotent.

Attributes

def fromValues(modelName: String, apiKey: String, baseUrl: String)(using resolver: ContextWindowResolver): Result[CohereConfig]

Constructs a CohereConfig, resolving contextWindow and reserveCompletion from the model name automatically.

Constructs a CohereConfig, resolving contextWindow and reserveCompletion from the model name automatically.

Value parameters

apiKey

Cohere API key; must be non-empty.

baseUrl

Compatibility API base, or a Cohere API root; must be non-empty. Mapped through compatibilityBaseUrl, and logged when that changes it. Defaults to CohereConfig.DEFAULT_BASE_URL when loaded via org.llm4s.config.Llm4sConfig.

modelName

Model identifier, e.g. "command-a-03-2025".

Attributes

Concrete fields

val COMPATIBILITY_PATH: String

The path of Cohere's OpenAI-compatibility API under an API root.

The path of Cohere's OpenAI-compatibility API under an API root.

Attributes

val DEFAULT_BASE_URL: String

Cohere's OpenAI-compatibility endpoint. This was https://api.cohere.com, the native API root, until the client moved onto the Compatibility API (#1132).

Cohere's OpenAI-compatibility endpoint. This was https://api.cohere.com, the native API root, until the client moved onto the Compatibility API (#1132).

Attributes

val DEFAULT_CONTEXT_WINDOW: Int

The context window apply uses when none is given; fromValues resolves the model's own.

The context window apply uses when none is given; fromValues resolves the model's own.

Attributes

val DEFAULT_RESERVE_COMPLETION: Int

The completion reserve apply uses when none is given; fromValues resolves the model's own.

The completion reserve apply uses when none is given; fromValues resolves the model's own.

Attributes

Generated with