llm4s-core
API
Generated with
llm4s-core/org.llm4s/org.llm4s.identity

org.llm4s.identity

package org.llm4s.identity

Members list

Type members

Classlikes

case class ModelId(name: String)

Identifies a specific LLM model by its API name string.

Identifies a specific LLM model by its API name string.

Model identifiers are forwarded verbatim to the provider's API. The companion object lists constants for known OpenAI models; other models (Anthropic, Gemini, Ollama, etc.) are referenced by passing the API name string directly to RuntimeId and the relevant org.llm4s.llmconnect.config.ProviderConfig.

Value parameters

name

Model identifier as expected by the provider API (e.g. "gpt-4o", "o3-mini").

Attributes

Companion
object
Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all
object ModelId

Attributes

Companion
class
Supertypes
trait Product
trait Mirror
class Object
trait Matchable
class Any
Self type
ModelId.type
case class RuntimeId(name: String)

Identifies an LLM runtime (provider backend) by its canonical name.

Identifies an LLM runtime (provider backend) by its canonical name.

Used to associate runtime-specific configuration and behaviour with a named backend, e.g. for tokenizer selection. Constants for all supported runtimes are defined on the companion object.

Value parameters

name

Canonical name string as expected by provider APIs (lower-case).

Attributes

Companion
object
Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all
object RuntimeId

Attributes

Companion
class
Supertypes
trait Product
trait Mirror
class Object
trait Matchable
class Any
Self type
RuntimeId.type
case class TokenizerId(name: String)

Identifies a BPE tokenizer vocabulary by its canonical name.

Identifies a BPE tokenizer vocabulary by its canonical name.

Tokenizer IDs are used by context-window estimation logic to select the correct byte-pair-encoding vocabulary for a given model, so that prompt and completion token counts are accurate without calling the provider API. The mapping from ModelId to TokenizerId is maintained by the context package.

Value parameters

name

Tokenizer vocabulary name as used by tiktoken and related libraries (e.g. "cl100k_base" for GPT-4 / GPT-3.5).

Attributes

Companion
object
Supertypes
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any
Show all
object TokenizerId

Attributes

Companion
class
Supertypes
trait Product
trait Mirror
class Object
trait Matchable
class Any
Self type
Generated with