llmrunnerpb

type ChatMessage struct {
	Id                int64   `protobuf:"varint,1,opt,name=id,proto3" json:"id,omitempty"`
	Content           string  `protobuf:"bytes,2,opt,name=content,proto3" json:"content,omitempty"`
	Role              string  `protobuf:"bytes,3,opt,name=role,proto3" json:"role,omitempty"`
	CreatedAt         int64   `protobuf:"varint,4,opt,name=created_at,json=createdAt,proto3" json:"created_at,omitempty"`
	AttachmentName    *string `protobuf:"bytes,5,opt,name=attachment_name,json=attachmentName,proto3,oneof" json:"attachment_name,omitempty"`
	AttachmentContent []byte  `protobuf:"bytes,6,opt,name=attachment_content,json=attachmentContent,proto3,oneof" json:"attachment_content,omitempty"`
	AttachmentMime    *string `protobuf:"bytes,7,opt,name=attachment_mime,json=attachmentMime,proto3,oneof" json:"attachment_mime,omitempty"`
	// contains filtered or unexported fields
}

type ChatResponse struct {
	Id               int64   `protobuf:"varint,1,opt,name=id,proto3" json:"id,omitempty"`
	Content          string  `protobuf:"bytes,2,opt,name=content,proto3" json:"content,omitempty"`
	Role             string  `protobuf:"bytes,3,opt,name=role,proto3" json:"role,omitempty"`
	CreatedAt        int64   `protobuf:"varint,4,opt,name=created_at,json=createdAt,proto3" json:"created_at,omitempty"`
	Done             bool    `protobuf:"varint,5,opt,name=done,proto3" json:"done,omitempty"`
	ReasoningContent *string `protobuf:"bytes,6,opt,name=reasoning_content,json=reasoningContent,proto3,oneof" json:"reasoning_content,omitempty"`
	PromptTokens     *int32  `protobuf:"varint,7,opt,name=prompt_tokens,json=promptTokens,proto3,oneof" json:"prompt_tokens,omitempty"`
	CompletionTokens *int32  `protobuf:"varint,8,opt,name=completion_tokens,json=completionTokens,proto3,oneof" json:"completion_tokens,omitempty"`
	TotalTokens      *int32  `protobuf:"varint,9,opt,name=total_tokens,json=totalTokens,proto3,oneof" json:"total_tokens,omitempty"`
	// contains filtered or unexported fields
}

type EmbedBatchRequest struct {
	Texts []string `protobuf:"bytes,1,rep,name=texts,proto3" json:"texts,omitempty"`
	// contains filtered or unexported fields
}

type EmbedBatchResponse struct {
	Embeddings []*Embedding `protobuf:"bytes,1,rep,name=embeddings,proto3" json:"embeddings,omitempty"`
	// contains filtered or unexported fields
}

type EmbedRequest struct {
	Text string `protobuf:"bytes,1,opt,name=text,proto3" json:"text,omitempty"`
	// contains filtered or unexported fields
}

type EmbedResponse struct {
	Values []float32 `protobuf:"fixed32,1,rep,packed,name=values,proto3" json:"values,omitempty"`
	// contains filtered or unexported fields
}

type Embedding struct {
	Values []float32 `protobuf:"fixed32,1,rep,packed,name=values,proto3" json:"values,omitempty"`
	// contains filtered or unexported fields
}

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

type GenerationParams struct {
	Temperature    *float32        `protobuf:"fixed32,1,opt,name=temperature,proto3,oneof" json:"temperature,omitempty"`
	MaxTokens      *int32          `protobuf:"varint,2,opt,name=max_tokens,json=maxTokens,proto3,oneof" json:"max_tokens,omitempty"`
	TopK           *int32          `protobuf:"varint,3,opt,name=top_k,json=topK,proto3,oneof" json:"top_k,omitempty"`
	TopP           *float32        `protobuf:"fixed32,4,opt,name=top_p,json=topP,proto3,oneof" json:"top_p,omitempty"`
	ResponseFormat *ResponseFormat `protobuf:"bytes,5,opt,name=response_format,json=responseFormat,proto3" json:"response_format,omitempty"`
	EnableThinking *bool           `protobuf:"varint,6,opt,name=enable_thinking,json=enableThinking,proto3,oneof" json:"enable_thinking,omitempty"`
	// contains filtered or unexported fields
}

type GetGpuInfoResponse struct {
	Gpus []*GpuInfo `protobuf:"bytes,1,rep,name=gpus,proto3" json:"gpus,omitempty"`
	// contains filtered or unexported fields
}

type GetLoadedModelResponse struct {
	Loaded       bool   `protobuf:"varint,1,opt,name=loaded,proto3" json:"loaded,omitempty"`
	GgufBasename string `protobuf:"bytes,2,opt,name=gguf_basename,json=ggufBasename,proto3" json:"gguf_basename,omitempty"`
	DisplayName  string `protobuf:"bytes,3,opt,name=display_name,json=displayName,proto3" json:"display_name,omitempty"`
	// contains filtered or unexported fields
}

type GetModelsResponse struct {
	Models []string `protobuf:"bytes,1,rep,name=models,proto3" json:"models,omitempty"`
	// contains filtered or unexported fields
}

type GpuInfo struct {
	Name               string `protobuf:"bytes,1,opt,name=name,proto3" json:"name,omitempty"`
	TemperatureC       int32  `protobuf:"varint,2,opt,name=temperature_c,json=temperatureC,proto3" json:"temperature_c,omitempty"`
	MemoryTotalMb      uint64 `protobuf:"varint,3,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"`
	MemoryUsedMb       uint64 `protobuf:"varint,4,opt,name=memory_used_mb,json=memoryUsedMb,proto3" json:"memory_used_mb,omitempty"`
	UtilizationPercent uint32 `protobuf:"varint,5,opt,name=utilization_percent,json=utilizationPercent,proto3" json:"utilization_percent,omitempty"`
	// contains filtered or unexported fields
}

type LLMRunnerServiceClient interface {
	// GetModels - имена моделей в model_path
	GetModels(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*GetModelsResponse, error)
	// GetServerInfo - hostname, платформа, число ядер, объём RAM
	GetServerInfo(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*ServerInfo, error)
	// GetGpuInfo - по каждой GPU: название, температура, VRAM, загрузка (%)
	GetGpuInfo(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*GetGpuInfoResponse, error)
	// RunnerProbe - агрегат для балансировщика: backend, хост, GPU, загруженная модель
	RunnerProbe(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*RunnerProbeResponse, error)
	// LoadModel - загрузить веса в RAM/VRAM; вызывать до SendMessage
	LoadModel(ctx context.Context, in *LoadModelRequest, opts ...grpc.CallOption) (*Empty, error)
	// GetLoadedModel - признак загрузки и идентификаторы
	GetLoadedModel(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*GetLoadedModelResponse, error)
	// UnloadModel - выгрузить текущую модель и освободить память
	UnloadModel(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*Empty, error)
	// ResetMemory - выгрузка модели (эквивалент UnloadModel)
	ResetMemory(ctx context.Context, in *Empty, opts ...grpc.CallOption) (*Empty, error)
	// SendMessage - server-stream чата; модель берётся из LoadModel
	SendMessage(ctx context.Context, in *SendMessageRequest, opts ...grpc.CallOption) (grpc.ServerStreamingClient[ChatResponse], error)
	// Embed - вектор одного текста
	Embed(ctx context.Context, in *EmbedRequest, opts ...grpc.CallOption) (*EmbedResponse, error)
	// EmbedBatch - векторы для списка текстов
	EmbedBatch(ctx context.Context, in *EmbedBatchRequest, opts ...grpc.CallOption) (*EmbedBatchResponse, error)
}

type LLMRunnerServiceServer interface {
	// GetModels - имена моделей в model_path
	GetModels(context.Context, *Empty) (*GetModelsResponse, error)
	// GetServerInfo - hostname, платформа, число ядер, объём RAM
	GetServerInfo(context.Context, *Empty) (*ServerInfo, error)
	// GetGpuInfo - по каждой GPU: название, температура, VRAM, загрузка (%)
	GetGpuInfo(context.Context, *Empty) (*GetGpuInfoResponse, error)
	// RunnerProbe - агрегат для балансировщика: backend, хост, GPU, загруженная модель
	RunnerProbe(context.Context, *Empty) (*RunnerProbeResponse, error)
	// LoadModel - загрузить веса в RAM/VRAM; вызывать до SendMessage
	LoadModel(context.Context, *LoadModelRequest) (*Empty, error)
	// GetLoadedModel - признак загрузки и идентификаторы
	GetLoadedModel(context.Context, *Empty) (*GetLoadedModelResponse, error)
	// UnloadModel - выгрузить текущую модель и освободить память
	UnloadModel(context.Context, *Empty) (*Empty, error)
	// ResetMemory - выгрузка модели (эквивалент UnloadModel)
	ResetMemory(context.Context, *Empty) (*Empty, error)
	// SendMessage - server-stream чата; модель берётся из LoadModel
	SendMessage(*SendMessageRequest, grpc.ServerStreamingServer[ChatResponse]) error
	// Embed - вектор одного текста
	Embed(context.Context, *EmbedRequest) (*EmbedResponse, error)
	// EmbedBatch - векторы для списка текстов
	EmbedBatch(context.Context, *EmbedBatchRequest) (*EmbedBatchResponse, error)
	// contains filtered or unexported methods
}

type LLMRunnerService_SendMessageClient = grpc.ServerStreamingClient[ChatResponse]

type LLMRunnerService_SendMessageServer = grpc.ServerStreamingServer[ChatResponse]

type LoadModelRequest struct {
	Model string `protobuf:"bytes,1,opt,name=model,proto3" json:"model,omitempty"`
	// contains filtered or unexported fields
}

type ResponseFormat struct {
	Type   string  `protobuf:"bytes,1,opt,name=type,proto3" json:"type,omitempty"`
	Schema *string `protobuf:"bytes,2,opt,name=schema,proto3,oneof" json:"schema,omitempty"`
	// contains filtered or unexported fields
}

type RunnerProbeResponse struct {
	BackendConnected bool                    `protobuf:"varint,1,opt,name=backend_connected,json=backendConnected,proto3" json:"backend_connected,omitempty"`
	Server           *ServerInfo             `protobuf:"bytes,2,opt,name=server,proto3" json:"server,omitempty"`
	Gpus             []*GpuInfo              `protobuf:"bytes,3,rep,name=gpus,proto3" json:"gpus,omitempty"`
	LoadedModel      *GetLoadedModelResponse `protobuf:"bytes,4,opt,name=loaded_model,json=loadedModel,proto3" json:"loaded_model,omitempty"`
	// contains filtered or unexported fields
}

type SendMessageRequest struct {
	Messages         []*ChatMessage    `protobuf:"bytes,1,rep,name=messages,proto3" json:"messages,omitempty"`
	StopSequences    []string          `protobuf:"bytes,2,rep,name=stop_sequences,json=stopSequences,proto3" json:"stop_sequences,omitempty"`
	TimeoutSeconds   *int32            `protobuf:"varint,3,opt,name=timeout_seconds,json=timeoutSeconds,proto3,oneof" json:"timeout_seconds,omitempty"`
	GenerationParams *GenerationParams `protobuf:"bytes,4,opt,name=generation_params,json=generationParams,proto3" json:"generation_params,omitempty"`
	RenderedPrompt   *string           `protobuf:"bytes,5,opt,name=rendered_prompt,json=renderedPrompt,proto3,oneof" json:"rendered_prompt,omitempty"`
	// contains filtered or unexported fields
}

type ServerInfo struct {
	Hostname      string `protobuf:"bytes,1,opt,name=hostname,proto3" json:"hostname,omitempty"`
	Os            string `protobuf:"bytes,2,opt,name=os,proto3" json:"os,omitempty"`
	Arch          string `protobuf:"bytes,3,opt,name=arch,proto3" json:"arch,omitempty"`
	CpuCores      int32  `protobuf:"varint,4,opt,name=cpu_cores,json=cpuCores,proto3" json:"cpu_cores,omitempty"`
	MemoryTotalMb uint64 `protobuf:"varint,5,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"`
	// contains filtered or unexported fields
}

type UnimplementedLLMRunnerServiceServer struct{}

type UnsafeLLMRunnerServiceServer interface {
	// contains filtered or unexported methods
}