ai_inference

Pydantic models for AI Inference service. These models are used for chat completions, text completions, and AI model management.

class ChatMessage(*, role: str, content: Optional[str] = None, name: Optional[str] = None, tool_calls: Optional[List[Dict[str, Any]]] = None, tool_call_id: Optional[str] = None) -> None

Chat message model for AI conversations.

class ChatRequest(*, model: Optional[str] = 'openai-gpt-4', messages: Annotated[List[datalayer_core.models.ai_inference.ChatMessage], MinLen(min_length=1)], temperature: Annotated[Optional[float], Ge(ge=0.0), Le(le=2.0)] = 0.7, max_tokens: Annotated[Optional[int], Ge(ge=1)] = 1024, stream: Optional[bool] = False, tools: Optional[List[Dict[str, Any]]] = None, tool_choice: Union[str, Dict[str, Any], NoneType] = None, functions: Optional[List[Dict[str, Any]]] = None, function_call: Union[str, Dict[str, Any], NoneType] = None, parallel_tool_calls: Optional[bool] = None) -> None

Chat completion request model.

class CompletionRequest(*, model: Optional[str] = 'text-model', prompt: str, temperature: Annotated[Optional[float], Ge(ge=0.0), Le(le=2.0)] = 0.7, max_tokens: Annotated[Optional[int], Ge(ge=1)] = 1024, stop: Optional[List[str]] = None, stream: Optional[bool] = False) -> None

Text completion request model.

class ChatResponseData(*, response: Optional[str] = None, message: Optional[Dict[str, Any]] = None, choices: Optional[List[Dict[str, Any]]] = None, model: Optional[str] = None, usage: Optional[Dict[str, Any]] = None) -> None

Chat completion response data model.

class CompletionResponseData(*, response: str, model: Optional[str] = None, usage: Optional[Dict[str, Any]] = None) -> None

Text completion response data model.

class ModelsResponseData(*, models: List[str], aliases: Dict[str, str] = <factory>, categories: Optional[Dict[str, List[str]]] = None) -> None

Available models response data model.

class HealthResponseData(*, status: str, service: str, version: Optional[str] = None, timestamp: Optional[str] = None) -> None

Health check response data model.

class EmbeddingRequest(*, model: Optional[str] = 'text-embedding-ada-002', input: Union[str, List[str]], encoding_format: Optional[str] = 'float', dimensions: Optional[int] = None) -> None

Embedding request model.

class EmbeddingData(*, object: str = 'embedding', embedding: List[float], index: int) -> None

Single embedding data model.

class EmbeddingResponseData(*, object: str = 'list', data: List[datalayer_core.models.ai_inference.EmbeddingData], model: str, usage: Optional[Dict[str, Any]] = None) -> None

Embedding response data model.