calute.api_server.models

calute.api_server.models#

Request and response models for the OpenAI-compatible API.

This module provides Pydantic models for the API server’s request and response handling. It includes: - Model information and listing responses - Health check response models - OpenAI-compatible data structures

All models follow the OpenAI API specification for compatibility with existing OpenAI client libraries and tools.

Example

>>> from calute.api_server.models import ModelInfo, ModelsResponse
>>> model = ModelInfo(id="my-agent", created=1234567890)
>>> response = ModelsResponse(data=[model])
class calute.api_server.models.HealthResponse(*, status: str, agents: int)[source]#

Bases: BaseModel

Health check response model.

Response format for the /health endpoint, providing basic health status information about the API server and its registered agents. Used by load balancers and monitoring systems to verify server availability.

status#

Health status string (e.g., "healthy", "degraded").

Type

str

agents#

Number of registered agents currently available for serving chat completion requests.

Type

int

Example

>>> from calute.api_server.models import HealthResponse
>>> health = HealthResponse(status="healthy", agents=3)
>>> health.status
'healthy'
agents: int#
model_config: ClassVar[ConfigDict] = {}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: str#
class calute.api_server.models.ModelInfo(*, id: str, object: str = 'model', created: int, owned_by: str = 'calute')[source]#

Bases: BaseModel

Information about an available model/agent.

Represents metadata for a single model or agent registered with the Calute API server. Follows the OpenAI model object specification for compatibility with the /v1/models endpoint.

id#

Unique identifier for the model/agent. This is the value clients use in the model field of chat completion requests.

Type

str

object#

Object type, always "model" for OpenAI compatibility.

Type

str

created#

Unix timestamp (seconds since epoch) indicating when the model entry was created.

Type

int

owned_by#

Owner identifier for the model. Defaults to "calute".

Type

str

Example

>>> from calute.api_server.models import ModelInfo
>>> info = ModelInfo(id="my-agent", created=1700000000)
>>> info.object
'model'
>>> info.owned_by
'calute'
created: int#
id: str#
model_config: ClassVar[ConfigDict] = {}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

object: str#
owned_by: str#
class calute.api_server.models.ModelsResponse(*, object: str = 'list', data: list[calute.api_server.models.ModelInfo])[source]#

Bases: BaseModel

Response containing a list of available models/agents.

Standard response format for the /v1/models endpoint, providing a list of all registered agents. Follows the OpenAI list response specification.

object#

Object type, always "list" for OpenAI compatibility.

Type

str

data#

List of ModelInfo objects representing all available agents and models registered with the server.

Type

list[calute.api_server.models.ModelInfo]

Example

>>> from calute.api_server.models import ModelInfo, ModelsResponse
>>> model = ModelInfo(id="assistant", created=1700000000)
>>> response = ModelsResponse(data=[model])
>>> response.object
'list'
>>> len(response.data)
1
data: list[calute.api_server.models.ModelInfo]#
model_config: ClassVar[ConfigDict] = {}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

object: str#
Contents