class LittleGhost::Model

Model is the resolved connection between an agent selection and a provider. It keeps provider behavior behind one executable interface whether an agent selected a role, canonical target, or inline configuration.

It merges profile settings into every ModelRequest, validates attachment modalities declared in metadata, lets providers prepare capability-sensitive requests, and delegates the normalized stream to the provider.

Attributes

Provider object, canonical target, default settings, normalized model details, and logical application role.

Provider object, canonical target, default settings, normalized model details, and logical application role.

Provider object, canonical target, default settings, normalized model details, and logical application role.

Provider object, canonical target, default settings, normalized model details, and logical application role.

Provider object, canonical target, default settings, normalized model details, and logical application role.

Public Class Methods

new (provider:, target:, settings: {}, details: nil, role: nil)

Source

# File lib/little_ghost/model.rb, line 34
def initialize(provider:, target:, settings: {}, details: nil, role: nil)
  raise ArgumentError, "provider must be a LittleGhost::Providers::Base" unless provider.is_a?(Providers::Base)

  @provider = provider
  @target = Models::Target.parse(target)
  @settings = settings.to_h.transform_keys(&:to_sym).freeze
  @role = role&.to_s
  @details = details || Models::Details.new(target: @target)
end

Connects a provider adapter to its canonical target, profile settings, optional model details, and logical role.

Public Instance Methods

capabilities ()

Source

# File lib/little_ghost/model.rb, line 100
def capabilities
  @capabilities ||= provider.capabilities(metadata: details.attributes)
end

Uses advertised provider capabilities.

embed (request)

Source

# File lib/little_ghost/model.rb, line 83
def embed(request)
  configured = Embeddings::Request.new(
    inputs: request.inputs,
    settings: settings.merge(request.settings),
    limits: request.limits,
    cancellation_token: request.cancellation_token,
    deadline: request.deadline
  )
  response = provider.embed(configured)
  unless response.is_a?(Embeddings::Response) && response.vectors.length == configured.inputs.length
    raise ProtocolError, "Embedding provider returned an unexpected vector count"
  end

  response
end

Applies profile settings and executes an Embeddings::Request.

Returns an Embeddings::Response with one vector per input. A provider that returns another shape raises ProtocolError.

Source

# File lib/little_ghost/model.rb, line 45
def model_id = target.model_id

Provider-owned identifier from the canonical target.

stream (request, &block)

Source

# File lib/little_ghost/model.rb, line 50
def stream(request, &block)
  validate_input_modalities!(request)
  configured_settings = settings.merge(request.settings)
  configured_max_tokens = if request.settings.key?(:max_tokens)
    request.settings[:max_tokens]
  elsif request.settings.key?("max_tokens")
    request.settings["max_tokens"]
  else
    settings[:max_tokens] || settings["max_tokens"]
  end
  if configured_max_tokens && details.max_output_tokens
    configured_settings.delete(:max_tokens)
    configured_settings.delete("max_tokens")
    configured_settings[:max_tokens] = [configured_max_tokens, details.max_output_tokens].min
  end
  configured_request = ModelRequest.new(
    messages: request.messages,
    tools: request.tools,
    settings: configured_settings,
    output_schema: request.output_schema,
    tool_choice: request.tool_choice,
    required_capabilities: request.required_capabilities,
    cancellation_token: request.cancellation_token,
    deadline: request.deadline
  )
  configured_request = provider.prepare_request(configured_request, capabilities:)
  provider.stream(configured_request, &block)
end

Streams request through the configured provider.

Profile settings are defaults; settings on request take precedence.