llms

type CallOption func(*CallOptions)

type CallOptions struct {
	// Model is the model to use.
	Model string `json:"model"`
	// MaxTokens is the maximum number of tokens to generate.
	MaxTokens int `json:"max_tokens,omitempty"`
	// Temperature is the temperature for sampling, between 0 and 1.
	Temperature float64 `json:"temperature,omitempty"`
	// StopWords is a list of words to stop on.
	StopWords []string `json:"stop_words"`
	// StreamingFunc is a function to be called for each chunk of a streaming response.
	// Return an error to stop streaming early.
	StreamingFunc func(ctx context.Context, chunk []byte) error
	// TopK is the number of tokens to consider for top-k sampling.
	TopK int `json:"top_k"`
	// TopP is the cumulative probability for top-p sampling.
	TopP float64 `json:"top_p"`
	// Seed is a seed for deterministic sampling.
	Seed int `json:"seed"`
	// MinLength is the minimum length of the generated text.
	MinLength int `json:"min_length"`
	// MaxLength is the maximum length of the generated text.
	MaxLength int `json:"max_length"`
	// N is how many chat completion choices to generate for each input message.
	N int `json:"n"`
	// RepetitionPenalty is the repetition penalty for sampling.
	RepetitionPenalty float64 `json:"repetition_penalty"`
	// FrequencyPenalty is the frequency penalty for sampling.
	FrequencyPenalty float64 `json:"frequency_penalty"`
	// PresencePenalty is the presence penalty for sampling.
	PresencePenalty float64 `json:"presence_penalty"`

	// Function defitions to include in the request.
	Functions []FunctionDefinition `json:"functions"`
	// FunctionCallBehavior is the behavior to use when calling functions.
	//
	// If a specific function should be invoked, use the format:
	// `{"name": "my_function"}`
	FunctionCallBehavior FunctionCallBehavior `json:"function_call"`
}

type ChatLLM interface {
	Call(ctx context.Context, messages []schema.ChatMessage, options ...CallOption) (*schema.AIChatMessage, error)
	Generate(ctx context.Context, messages [][]schema.ChatMessage, options ...CallOption) ([]*Generation, error)
}

type FunctionCallBehavior string

type FunctionDefinition struct {
	// Name is the name of the function.
	Name string `json:"name"`
	// Description is a description of the function.
	Description string `json:"description"`
	// Parameters is a list of parameters for the function.
	Parameters any `json:"parameters"`
}

type Generation struct {
	// Text is the generated text.
	Text string `json:"text"`
	// Message stores the potentially generated message.
	Message *schema.AIChatMessage `json:"message"`
	// GenerationInfo is the generation info. This can contain vendor-specific information.
	GenerationInfo map[string]any `json:"generation_info"`
}

type LLM interface {
	Call(ctx context.Context, prompt string, options ...CallOption) (string, error)
	Generate(ctx context.Context, prompts []string, options ...CallOption) ([]*Generation, error)
}

type LLMResult struct {
	Generations [][]*Generation
	LLMOutput   map[string]any
}

type LanguageModel interface {
	// Take in a list of prompt values and return an LLMResult.
	GeneratePrompt(ctx context.Context, prompts []schema.PromptValue, options ...CallOption) (LLMResult, error)
	// Get the number of tokens present in the text.
	GetNumTokens(text string) int
}