llm

type CompletionRequest struct {
	Prompt  string
	Format  string
	Images  []ImageData
	Options api.Options
}

type CompletionResponse struct {
	Content            string
	Done               bool
	PromptEvalCount    int
	PromptEvalDuration time.Duration
	EvalCount          int
	EvalDuration       time.Duration
}

type DetokenizeRequest struct {
	Tokens []int `json:"tokens"`
}

type DetokenizeResponse struct {
	Content string `json:"content"`
}

type EmbeddingRequest struct {
	Content string `json:"content"`
}

type EmbeddingResponse struct {
	Embedding []float64 `json:"embedding"`
}

type GGML struct {
	// contains filtered or unexported fields
}

type ImageData struct {
	Data []byte `json:"data"`
	ID   int    `json:"id"`
}

type KV map[string]any

type Layer map[string]*Tensor

type LlamaServer interface {
	Ping(ctx context.Context) error
	WaitUntilRunning(ctx context.Context) error
	Completion(ctx context.Context, req CompletionRequest, fn func(CompletionResponse)) error
	Embedding(ctx context.Context, prompt string) ([]float64, error)
	Tokenize(ctx context.Context, content string) ([]int, error)
	Detokenize(ctx context.Context, tokens []int) (string, error)
	Close() error
	EstimatedVRAM() uint64
}

type ServerStatus int

type ServerStatusResp struct {
	Status          string `json:"status"`
	SlotsIdle       int    `json:"slots_idle"`
	SlotsProcessing int    `json:"slots_processing"`
	Error           string `json:"error"`
}

type StatusWriter struct {
	LastErrMsg string
	// contains filtered or unexported fields
}

type Tensor struct {
	Name   string `json:"name"`
	Kind   uint32 `json:"kind"`
	Offset uint64 `json:"-"`

	// Shape is the number of elements in each dimension
	Shape []uint64 `json:"shape"`

	io.WriterTo `json:"-"`
}

type Tensors []*Tensor

type TokenizeRequest struct {
	Content string `json:"content"`
}

type TokenizeResponse struct {
	Tokens []int `json:"tokens"`
}